Daily reading queue

150 items for 2026-06-05 across 4 categories.

Active sources: 7. Sources represented in this queue: 6. The cron runs daily at 06:00 server time; arxiv RSS is often empty on weekends.

Linked to your results

score 100arxiv cs.LG (Machine Learning)arxiv:2606.04176unread
Low-rank Distributional Matrix Completion
Jiayi Wang, Raymond K. W. Wong · 2026-06-05
arXiv:2606. 04176v1 Announce Type: new Abstract: We study a distributional generalization of the matrix completion problem in which each entry of the target matrix is a probability distribution rather than a scalar.
Read next because Low-rank Distributional Matrix Completion overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, rect, under, distributional, rate, completion. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04176v1 Announce Type: new Abstract: We study a distributional generalization of the matrix completion problem in which each entry of the target matrix is a probability distribution rather than a scalar. In this setting, only a subset of matrix entries is observed, and even for observed entries, the underlying distributions are not directly accessible; instead, we observe finitely many samples drawn from them. To represent distributional entries, we employ kernel mean embeddings and introduce a notion of Tucker rank for distribution-valued matrices to capture their low-rank structure. The infinite-dimensional nature of kernel embeddings poses significant methodological challenges. To address this, we introduce functional unfolding operators that link the proposed distributional low-rank structure to the classical Tucker rank for finite-dimensional tensors. Based on this framework, we propose a novel estimator for distributional matrix completion. We establish non-asymptotic error bounds that characterize the statistical performance of the estimator. Extensive experiments on synthetic data and a real-world application demonstrate the effectiveness of the proposed method.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04167unread
Smart Transportation Without Neurons -- Fair Metro Network Expansion with Tabular Reinforcement Learning
Dimitris Michailidis, Sennay Ghebreab, Fernando P. Santos · 2026-06-05
arXiv:2606. 04167v1 Announce Type: new Abstract: We tackle the Metro Network Expansion Problem (MNEP), a subset of the Transport Network Design Problem (TNDP), which focuses on expanding metro systems to satisfy travel demand.
Read next because Smart Transportation Without Neurons -- Fair Metro Network Expansion with Tabular Reinforcement Learning overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: latin, eval, source, rate, without, factor. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04167v1 Announce Type: new Abstract: We tackle the Metro Network Expansion Problem (MNEP), a subset of the Transport Network Design Problem (TNDP), which focuses on expanding metro systems to satisfy travel demand. Traditional methods rely on exact and heuristic approaches that require expert-defined constraints to reduce the search space. Recently, deep reinforcement learning (Deep RL) has emerged due to its effectiveness in complex sequential decision-making processes-it remains, however, computationally expensive, environmentally costly, and requires additional engineering to interpret. We show that MNEP problems are small enough to not require Deep RL methods. Reformulating the MNEP as a Non-Markovian Rewards Decision Process (NMRDP), we use tabular RL to achieve similar performance with significantly fewer training episodes, additionally offering greater interpretability. Additionally, we incorporate social equity criteria into the reward functions, focusing on efficiency and fairness, highlighting the versatility of our method. Evaluated in real-world settings-Xi'an and Amsterdam-our method reduces total episodes by a factor of 18 and total carbon emissions by a factor of 12 on average, while remaining competitive with Deep RL. This approach offers a replicable, modular, interpretable, and resource-efficient solution with potential applications to other combinatorial optimization problems.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04161unread
When Offline Selectors Cannot Beat the Best Single Model: A Diagnostic Study on edX Dropout Prediction
Tyler Crosse, Alan Nadelsticher Ruvalcaba, Dustin Khang LeDuc, Thomas Trask, Nicholas Lytle, David Joyner · 2026-06-05
arXiv:2606. 04161v1 Announce Type: new Abstract: Different predictors often excel on different inputs, so picking the best one per instance promises higher accuracy than committing to a single model.
Read next because When Offline Selectors Cannot Beat the Best Single Model: A Diagnostic Study on edX Dropout Prediction overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, line, rate, without, does, trained, sweep, stage. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04161v1 Announce Type: new Abstract: Different predictors often excel on different inputs, so picking the best one per instance promises higher accuracy than committing to a single model. In practice, selectors trained from logged data routinely fail to beat the strongest single predictor. Three causes typically go unseparated before more tuning is applied: a mismatched learner, a state that does not predict which model wins, or buffer-to-deployment label shift. A three-stage diagnostic rules them out on a shared buffer. Stage~1 estimates a local ceiling on oracle recovery from $k$-NN label consistency. Stage~2 asks whether paired BC and offline-RL learners (BC, DQN, and CQL across penalty weights) reach that ceiling. Stage~3 ablates the selector state to test whether richer features would raise it. The combined verdict points to the most promising next step: tuning the learner, redesigning the state, or collecting new data. We apply it to selecting among five dropout-prediction models on edX clickstream data. Across 16 windows, the oracle beats the strongest single base model by 9.7 accuracy points on average, yet BC, DQN, and CQL land in the same test-accuracy band below it (robust to a tenfold buffer sweep and $N{=}2{,}000$ held-out examples). The bottleneck is local representational ambiguity: CQL closes the imitation gap without a deployment gain (not conservatism), regret clusters tightly across learners (not tie-breaking), and the three learners converge on test accuracy (not shift). The next iteration should change the state or collect new data, not tune the offline learner further.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04145unread
EvalStop: Using World Feedback to Detect and Correct Reward Overoptimization in Multi-Tenant RLHF Platforms
Guilin Zhang, Chuanyi Sun, Shahryar Sarkani, John M. Fossaceca · 2026-06-05
arXiv:2606. 04145v1 Announce Type: new Abstract: Cloud LLM fine-tuning platforms increasingly serve RLHF workloads, where a learned reward model is optimized as a proxy for human quality.
Read next because EvalStop: Using World Feedback to Detect and Correct Reward Overoptimization in Multi-Tenant RLHF Platforms overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: class, rect, under, correct, eval, line, rate, without. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04145v1 Announce Type: new Abstract: Cloud LLM fine-tuning platforms increasingly serve RLHF workloads, where a learned reward model is optimized as a proxy for human quality. As Gao et al. (2023) showed, this proxy diverges from world feedback (downstream eval metrics) under sustained optimization pressure, a phenomenon known as reward overoptimization. Existing platform schedulers ignore this divergence: non-clairvoyant schedulers optimize JCT without any quality signal, SLAQ-style quality-aware schedulers use training loss (a weaker proxy that drops monotonically through hacking), and classical per-job early stopping requires human monitoring and does not free shared GPUs. We propose EvalStop, a composable scheduling primitive that terminates jobs on k consecutive eval-score declines, releases GPUs, preserves the best checkpoint, and delegates to any base scheduler. We frame scheduler-level early stopping as a detection problem and evaluate it in a discrete-event simulator whose RLHF workload mixes reward-hacking and structurally healthy runs, with ground-truth labels hidden from schedulers. On RLHF-heavy workloads (80% RLHF, 64 GPUs), EvalStop achieves precision 98% / recall 99% / FPR 1.5% while improving JCT by 9% and cutting wasted compute by 22% over SRTF-Est (p<0.05). Trivial fixed-progress and loss-plateau competitors either incur 65% FPR on healthy RLHF or miss over half of true hacking cases. Gains compose across every base scheduler tested (9-25% JCT) and detection quality stays stable under eval noise (precision at least 91% at noise std <= 0.05) and hacking base rate (precision at least 89% across 20-80% hacking fractions).
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04100unread
Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials
Joanna Zou, Fraser Birks, Dallas Foster, Youssef Marzouk · 2026-06-05
arXiv:2606. 04100v1 Announce Type: new Abstract: Machine learning interatomic potentials (MLIPs) enable efficient and accurate atomistic simulations but depend critically on the quality and diversity of the training data.
Read next because Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Implement Chen et al. persona-vector extraction recipe and compare to project's centroid-difference recipe", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: line, rate, compare, symmetry, lora, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04100v1 Announce Type: new Abstract: Machine learning interatomic potentials (MLIPs) enable efficient and accurate atomistic simulations but depend critically on the quality and diversity of the training data. We introduce Stein kernelized molecular dynamics (SKMD), an enhanced sampling method that uses interacting particle dynamics to acquire informative training configurations for the active learning and fine-tuning of MLIPs. SKMD corresponds to a stochastic variant of Stein variational gradient descent that is adapted for molecular dynamics by incorporating asynchronous particle updates and a kernel of global atomic descriptors, which provides a symmetry-aware measure of configurational similarity. Unlike other enhanced samplers used in molecular dynamics, SKMD preserves the Boltzmann distribution as the asymptotic distribution of the dynamics. This property enforces a balance between the exploration of diverse configurations and attraction toward high-probability regions of the energy landscape. We further propose an approach to efficient online data acquisition using an adaptive stopping criterion that selects non-redundant training data over the course of simulation. We demonstrate SKMD for the active learning of a neural network model of the M\"uller-Brown potential and the fine-tuning of a MACE interatomic potential for alanine dipeptide. Compared to active learning baselines, our method achieves higher model accuracy in fewer training iterations with the same number of acquired training samples.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04050unread
LiftQuant: Continuous Bit-Width LLM via Dimensional Lifting and Projection
Liulu He, XuanAng Liu, Juntao Liu, Taolue Feng, Ting Lu, Chunsheng Gan, Zhiyv Peng, Yuan Du, Huanrui Yang, Yijiang Liu, Li Du · 2026-06-05
arXiv:2606. 04050v1 Announce Type: new Abstract: Existing quantization methods are fundamentally limited by rigid, integer-based bit-widths (e.
Read next because LiftQuant: Continuous Bit-Width LLM via Dimensional Lifting and Projection overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, width, line, rate, project, control, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04050v1 Announce Type: new Abstract: Existing quantization methods are fundamentally limited by rigid, integer-based bit-widths (e.g., 2, 3-bit), resulting in a ``deployment gap" where Large Language Models cannot be optimally fitted to specific memory budgets. To bridge this gap, we introduce LiftQuant, a novel framework that enables continuous bit-width control for true Pareto-optimal deployment. The core innovation is a ``lift-then-project" mechanism which approximates low-dimensional weight vectors by projecting a simple 1-bit lattice from a higher-dimensional ``lifted" space. Crucially, the effective bit-width is determined simply by the ratio of the lifted dimension to the original dimension, which allows the bit-width to be tuned quasi-continuous as the dimension is a flexible structural parameter. This projection generates a structured yet non-uniform codebook, capturing the expressive power of Vector Quantization (VQ). While beneficial over VQ, LiftQuant's decoding path relies solely on linear transformations and 1-bit uniform quantizers, retaining hardware-friendly nature. This flexibility is transformative: LiftQuant enables a 70B LLM to be compressed to 2.4 bits to precisely fit a 24GB GPU, where its performance significantly surpasses state-of-the-art 2-bit models fitted on the same device. Our code and ckpt is available at https://github.com/Heliulu/LiftQuant.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04048unread
Unlocking Feature Learning in Gated Delta Networks at Scale
Yifeng Liu, Quanquan Gu · 2026-06-05
arXiv:2606. 04048v1 Announce Type: new Abstract: Training and scaling Large Language Models demand enormous computational resources, motivating both efficient sub-quadratic architectures and principled hyperparameter tuning methods.
Read next because Unlocking Feature Learning in Gated Delta Networks at Scale overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: rect, under, width, correct, source, line, rate, language. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04048v1 Announce Type: new Abstract: Training and scaling Large Language Models demand enormous computational resources, motivating both efficient sub-quadratic architectures and principled hyperparameter tuning methods. While the Maximal Update Parametrization ($\mu$P) has enabled zero-shot hyperparameter transfer for standard Transformers, its extension to linear models, particularly those with structured state transitions and complicated architectures, remains largely unexplored. By rigorously propagating coordinate-size estimates through the forward pass, gating mechanisms, and recurrent state dynamics, we derive the scaling rules for Gated Delta Network. Experiments on language-model pre-training confirm that our configurations enable stable learning-rate transfer across model widths under both AdamW and SGD, whereas standard parametrization fails to transfer, validating the correctness and practical utility of our analysis.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04045unread
Bayes-Sufficient Representations in Supervised Learning
Vasileios Sevetlidis · 2026-06-05
arXiv:2606. 04045v1 Announce Type: new Abstract: Representation learning is often described as preserving the information in an input that is relevant for prediction.
Read next because Bayes-Sufficient Representations in Supervised Learning overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: class, rate, implement, control. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04045v1 Announce Type: new Abstract: Representation learning is often described as preserving the information in an input that is relevant for prediction. This work asks what relevance means for a fixed supervised decision problem. A representation is defined to be Bayes-sufficient for a joint distribution and loss if some prediction head can use it to implement a Bayes-optimal action rule. This makes the target information loss-dependent. In the almost-surely unique Bayes-action case, the relevant object is a Bayes quotient, which identifies inputs that require the same Bayes-optimal action. A representation is sufficient when it refines this quotient, and Bayes-minimal when it is informationally equivalent to it. The framework connects naturally to property elicitation: zero-one loss requires the Bayes class, squared loss the conditional mean, Brier loss the conditional probability in binary prediction, and log loss or strictly proper scoring rules the predictive distribution. Controlled finite experiments, learned neural bottleneck experiments, and a real-data iNaturalist taxonomic refinement experiment illustrate the distinction between sufficiency, minimality, and retained non-required information. For a fixed supervised problem, the distribution and the loss determine the Bayes action, the Bayes action determines the quotient, and the quotient determines the minimal information required for Bayes-optimal prediction.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04036unread
Self-Distilled Policy Gradient
Yifeng Liu, Shiyuan Zhang, Yifan Zhang, Quanquan Gu · 2026-06-05
arXiv:2606. 04036v1 Announce Type: new Abstract: On-policy self-distillation, where a language model conditions on privileged context to supervise its own generations, is a promising source of dense supervision for sparse-reward reinforcement learning.
Read next because Self-Distilled Policy Gradient overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, text, source, line, full, on-policy, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04036v1 Announce Type: new Abstract: On-policy self-distillation, where a language model conditions on privileged context to supervise its own generations, is a promising source of dense supervision for sparse-reward reinforcement learning. Actually, it can be instantiated as an auxiliary full-vocabulary student-to-teacher reverse Kullback-Leibler divergence loss. We therefore propose SDPG, a self-distilled policy-gradient framework that combines group-relative verifier advantages with normalized standard deviation, exact full-vocabulary on-policy self-distillation, as well as reference-policy KL regularization. Empirically, SDPG improves stability and performance over RLVR and self-distillation baselines. The code is available at https://github.com/lauyikfung/SDPG.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04032unread
Do Transformers Need Three Projections? Systematic Study of QKV Variants
Ali Kayyam, Anusha Madan Gopal, M Anthony Lewis · 2026-06-05
arXiv:2606. 04032v2 Announce Type: new Abstract: Transformers have become the standard solution for various AI tasks, with the query, key, and value (QKV) attention formulation playing a central role.
Read next because Do Transformers Need Three Projections? Systematic Study of QKV Variants overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, under, eval, token, rate, project, position. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04032v2 Announce Type: new Abstract: Transformers have become the standard solution for various AI tasks, with the query, key, and value (QKV) attention formulation playing a central role. However, the individual contribution of these three projections and the impact of omitting some remain poorly understood. We systematically evaluate three projection sharing constraints: a) Q-K=V (shared key-value), b) Q=K-V (shared query-key), and c) Q=K=V (single projection). The last two variants produce symmetric attention maps; to address this, we also explore asymmetric attention via 2D positional encodings. Through experiments spanning synthetic tasks, vision (MNIST, CIFAR, TinyImageNet, anomaly), and language modeling (300M and 1.2B parameter models on 10B tokens), we discovered that our transformers perform on par or occasionally better than the QKV transformer. In language modeling, Q-K=V projection sharing achieves 50% KV cache reduction with only 3.1% perplexity degradation. Crucially, projection sharing is complementary to head sharing (GQA/MQA): combining Q-K=V with GQA-4 yields 87.5% cache reduction, while Q-K=V + MQA achieves 96.9%, enabling practical on-device inference. We show that Q-K=V preserves quality because keys and values can occupy similar representational spaces and attention operates in a low-rank regime, whereas Q=K-V breaks attention directionality. Our results systematically characterize projection sharing as an underexplored instance of weight tying in attention, with direct, quantifiable inference memory benefits, particularly valuable for edge deployment. The code is publicly available at https://github.com/Brainchip-Inc/Do-Transformers-Need-3-Projections
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04029unread
Position: Deployed Reinforcement Learning should be Continual
Parnian Behdin, Kevin Roice, Golnaz Mesbahi · 2026-06-05
arXiv:2606. 04029v1 Announce Type: new Abstract: Reinforcement Learning (RL) has received increasing attention and adoption in real-world use cases.
Read next because Position: Deployed Reinforcement Learning should be Continual overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?". Matching terms: eval, source, trained, position, never. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04029v1 Announce Type: new Abstract: Reinforcement Learning (RL) has received increasing attention and adoption in real-world use cases. Most of these systems follow a train-then-fix paradigm, where trained agents do not learn while interacting with the world until performance degrades and retraining becomes necessary. In this position paper, we argue that deploying an agent that is incapable of optimality, but receives an evaluative reward signal, is inherently a continual RL problem. We identify four sources of non-stationarity after deployment that necessitate never-ending learning, and highlight why the best deployed agents never stop adapting. We analyze successful examples of continual RL in the real world, and present the community with the advantages and measures to move away from the current train-then-fix paradigm.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.03995unread
Early Detection of Alzheimer's Disease Using Explainable Machine Learning on Clinical Biomarkers: A Multi-Class Classification Study Using the Alzheimer's Disease Neuroimaging Initiative (ADNI) Dataset
Afshan Hashmi · 2026-06-05
arXiv:2606. 03995v1 Announce Type: new Abstract: Background: Alzheimer's disease (AD) affects over 55 million people worldwide.
Read next because Early Detection of Alzheimer's Disease Using Explainable Machine Learning on Clinical Biomarkers: A Multi-Class Classification Study Using the Alzheimer's Disease Neuroimaging Initiative (ADNI) Dataset overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: marker, class, eval, line, rate, trained, test, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.03995v1 Announce Type: new Abstract: Background: Alzheimer's disease (AD) affects over 55 million people worldwide. Accurate, interpretable detection of normal cognition (NC), mild cognitive impairment (MCI), and AD from routine clinical assessments remains a critical unmet need. Methods: An XGBoost classifier was developed for three-class detection using eight clinical features from the Alzheimer's Disease Neuroimaging Initiative (ADNI): MMSE, CDR Global, CDR Sum of Boxes (CDR-SB), MoCA, FAQ, age, sex, and education. Hyperparameters were optimised using Optuna (50 trials); class imbalance was addressed with SMOTE. Performance was evaluated by macro AUC-ROC with 1,000-iteration bootstrap 95% confidence intervals, macro F1, balanced accuracy, and Cohen's kappa. SHAP values provided feature-level explainability. Results: The dataset comprised 1,641 baseline subjects (608 NC, 767 MCI, 266 AD). On five-fold cross-validation, mean macro AUC was 0.983 (SD 0.007), accuracy 0.944 (SD 0.006), and macro F1 0.929 (SD 0.008). On the held-out test set (n = 247), macro AUC was 0.982 (95% CI: 0.965--0.995), accuracy 0.943, balanced accuracy 0.932, macro F1 0.927, and Cohen's kappa 0.909. SHAP analysis identified CDR Global as the dominant predictor for NC and MCI, while CDR-SB and MMSE together drove AD classification. Conclusion: An explainable machine learning model trained on routine clinical assessments achieves near-perfect three-class Alzheimer's detection. SHAP analysis reveals clinically plausible, class-specific feature importance patterns supporting clinical validity. Future work will extend this framework with speech biomarkers for multimodal detection.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05560unread
Wasserstein Exponential Smoothing
Takuo Matsubara, Peiwen Jiang, Minh-Ngoc Tran, Wilson Ye Chen · 2026-06-05
arXiv:2606. 05560v1 Announce Type: cross Abstract: Exponential smoothing (ES) often outperforms other techniques in time series forecasting across a wide range of data-generating processes.
Read next because Wasserstein Exponential Smoothing overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, distributional, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05560v1 Announce Type: cross Abstract: Exponential smoothing (ES) often outperforms other techniques in time series forecasting across a wide range of data-generating processes. While ES has traditionally been applied to time series in $\mathbb{R}$, this paper extends the methodology to distributional time series, where each observation is a probability distribution on $\mathbb{R}$. The primary contribution of this work is twofold. First, we propose a principled and intuitive generalization of ES within the Wasserstein space, which retains the exceptional parsimony of classical ES. Second, we theoretically and empirically demonstrate that the smoothing parameter can be consistently estimated by minimizing a Wasserstein distance. Applications to distributional time series of high-frequency financial returns and household electricity demands confirm the practical effectiveness of our Wasserstein ES model.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05462unread
A Two-Channel F-Transform Representation for Early Trajectory Characterization in Iterated Correlation Dynamics
Ishrak Alhajj Hassan · 2026-06-05
arXiv:2606. 05462v1 Announce Type: cross Abstract: Many nonlinear iterative procedures generate high-dimensional trajectories whose early behavior is informative but difficult to compare directly.
Read next because A Two-Channel F-Transform Representation for Early Trajectory Characterization in Iterated Correlation Dynamics overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: rect, soft, line, rate, compare, project, length, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05462v1 Announce Type: cross Abstract: Many nonlinear iterative procedures generate high-dimensional trajectories whose early behavior is informative but difficult to compare directly. This paper studies a soft-computing representation problem: how to convert a short early trajectory segment into compact, interpretable, fixed-dimensional fuzzy coordinates that preserve information about subsequent convergence and trajectory geometry. The problem is investigated for iterated Pearson correlation matrices, a nonlinear matrix iteration historically connected with CONCOR-type blockmodeling and repeated-correlation methods. The proposed descriptor uses two logarithmic signals from the early post-transient regime: a step-size signal, measuring contraction magnitude, and a contraction-ratio signal, measuring local contraction evolution. Each signal is projected onto a three-node triangular fuzzy partition using zero-degree F-transform coefficients and one centered first-degree coefficient. This yields an eight-dimensional two-channel representation separating local level from local trend and contraction magnitude from contraction evolution. Across 22 matrix dimensions with 1000 trajectories per dimension, the descriptor is compared with raw trajectory samples, statistical summaries, and PCA-compressed raw features using Random Forest regression for convergence-length approximation. It achieves mean R^2 = 0.6480, close to raw trajectories (0.6518) and statistical summaries (0.6528), while improving over the step-size-only F-transform descriptor (0.5001). Repeated random-split and shifted-window experiments confirm stability. PCA and clustering further show reproducible low-dimensional organization, with the first two principal components explaining 84.26% of variance and k = 3 favored by the mean silhouette criterion.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05335unread
A prism hierarchy of learning regimes in large linear autoencoders
Eugene Golikov, Yaroslav Gusev, Dmitry Yarotsky · 2026-06-05
arXiv:2606. 05335v1 Announce Type: cross Abstract: Theoretical studies of machine learning models commonly consider different limiting regimes in which the learning dynamics of gradient descent becomes theoretically tractable.
Read next because A prism hierarchy of learning regimes in large linear autoencoders overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, under, good, line, does, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05335v1 Announce Type: cross Abstract: Theoretical studies of machine learning models commonly consider different limiting regimes in which the learning dynamics of gradient descent becomes theoretically tractable. It is, however, desirable to have a systematically obtained picture of all qualitatively different extreme learning regimes for a particular type of models. In this paper we propose such a picture for large weight-tied linear autoencoders characterized by input and latent dimensions, initialization magnitude, and training set size. This model is nonlinear in the weights and its gradient flow does not have a general theoretical solution. We show that at the level of the formal loss-expansion hierarchy, its extreme regimes are naturally associated with faces of a triangular prism. In particular, there are five basic extreme regimes associated with the 2-faces of the prism: (1) large-data, (2) small-data, (3) mean-field, (4) narrow-latent, and (5) free. For regimes (1,2,3,4), we derive explicit expressions for both train and population limiting loss evolutions under gradient flow, obtaining very good agreement with experimental results.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05327unread
Multimarginal flow matching with optimal transport potentials
Raghav Kansal, David Crair, Nghia Nguyen, Scott Pope, Bradley Parry · 2026-06-05
arXiv:2606. 05327v1 Announce Type: cross Abstract: Flow matching (FM) has emerged as a powerful framework for learning dynamic transport maps between two empirical distributions.
Read next because Multimarginal flow matching with optimal transport potentials overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, soft, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05327v1 Announce Type: cross Abstract: Flow matching (FM) has emerged as a powerful framework for learning dynamic transport maps between two empirical distributions. However, less explored is the setting with intermediate observed marginals that can help constrain the flows between the endpoints. This "multimarginal" regime is central to modeling temporal evolution in dynamical systems in many scientific domains that can sample sequential distributions. We tackle this problem with a novel approach that leverages the connection between FM and dynamic optimal transport (OT), softly steering the flow towards the intermediate marginals through potential terms in the dynamic OT action. By extending the conditional FM learning target to incorporate these potentials, we derive an efficient, simulation-free algorithm for multimarginal FM that offers considerable flexibility in the spatiotemporal dynamics of the learned flows. We demonstrate state-of-the-art performance and training efficiency of OT-potential FM (OTP-FM) on diverse single-cell RNA sequencing, oceanographic, and meteorological datasets. Our code is available at https://github.com/Bexorg-Inc/OTP-FM.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05247unread
DiffSlack: Learning under Nonlinear Inequality Constraints via Learnable Slack Variables
Ziqian Wang, Chenxi Fang, Zhen Zhang · 2026-06-05
arXiv:2606. 05247v1 Announce Type: cross Abstract: Enforcing nonlinear inequality constraints in neural networks remains challenging, especially when the output is subject to many coupled constraints.
Read next because DiffSlack: Learning under Nonlinear Inequality Constraints via Learnable Slack Variables overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, under, eval, line, rate, compare, project, trained. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05247v1 Announce Type: cross Abstract: Enforcing nonlinear inequality constraints in neural networks remains challenging, especially when the output is subject to many coupled constraints. Existing hard constraint methods often impose structural restrictions on the constraint set or introduce substantial computational overhead for large-scale nonlinear problems. Here, we propose DiffSlack, a differentiable projection layer for nonlinear inequality-constrained neural prediction. DiffSlack reformulates inequalities as equalities with learnable slack variables, which are predicted as part of the augmented network output and provide a data-driven warm start for damped Gauss-Newton projection. The projection layer maps raw predictions onto the augmented feasible manifold while preserving end-to-end differentiability. A two-stage curriculum further stabilizes training and improves constraint satisfaction. We evaluate DiffSlack on vehicle path planning with 200 nonlinear inequality constraints from collision avoidance, curvature limits, and waypoint spacing. Compared with existing learning-based baselines, DiffSlack achieves a higher planning success rate and stronger geometric constraint satisfaction under a comparable inference budget. Ablation studies further show that the hard projection layer reduces sensitivity to supervision quality. Closed-loop tracking in CARLA and real-world vehicle experiments confirms the executability of the generated trajectories. These results demonstrate that DiffSlack provides a practical and scalable approach to embedding hard inequality constraints into neural networks for engineering applications.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.06233unread
Anchor PCA
Benedikt Seiter, Anya Fries, Julius von K\"ugelgen, Jonas Peters · 2026-06-05
arXiv:2606. 06233v1 Announce Type: new Abstract: Principal component analysis (PCA) is one of the most widely used unsupervised dimension reduction techniques.
Read next because Anchor PCA overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, under, line, rate. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06233v1 Announce Type: new Abstract: Principal component analysis (PCA) is one of the most widely used unsupervised dimension reduction techniques. We study PCA for data from multiple related domains. Since principal components generally differ across domains, one way to obtain a shared low-rank embedding is to perform PCA on the pooled data. However, this approach can focus on spurious directions that exhibit high variation in only a few domains. To find a robust embedding that still explains most variance in unseen but similar domains, we propose instead to focus on shared directions of variation. To this end, we introduce Anchor PCA which trades off overall explained variance with agreement between the shared and domain-specific low-rank embeddings. Anchor PCA amounts to PCA on a modified target matrix and thus can be solved efficiently. Moreover, we show that Anchor PCA recovers a maximal invariant subspace and admits a minimax reconstruction interpretation under bounded domain-specific covariance inflations. On simulated and real-world gas sensor data with temporal drift, we demonstrate, respectively, that Anchor PCA recovers the maximally invariant subspace and yields embeddings that explain more variance on unseen domains than the pooling baseline and a worst-case alternative. Taken together, these findings establish Anchor PCA as a promising approach to robust unsupervised dimension reduction from multi-domain data.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.06179unread
Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
Na\"il B. Khelifa, Richard E. Turner, Ramji Venkataramanan · 2026-06-05
arXiv:2606. 06179v1 Announce Type: new Abstract: Score-based diffusion models are typically trained by minimizing the $L^2$ score matching error, and standard theoretical analyses rely on this quantity to bound the sampling discrepancy between the learned and target distributions.
Read next because Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, correct, distributional, full, trained, position, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06179v1 Announce Type: new Abstract: Score-based diffusion models are typically trained by minimizing the $L^2$ score matching error, and standard theoretical analyses rely on this quantity to bound the sampling discrepancy between the learned and target distributions. We show the $L^2$ score error is not the right intrinsic measure of marginal distributional quality: a learned diffusion model can incur arbitrarily large $L^2$ score error while perfectly matching the target distribution. By decomposing score errors into a gradient and a solenoidal component (a Helmholtz-Hodge decomposition), we identify the geometric reason behind this: only the gradient component enters the marginal Fokker-Planck dynamics, while the solenoidal component is structurally invisible. We make this precise in three results. First, building on the corrected geometry, we prove an impossibility result: no monotone function of the $L^2$ score error can uniformly lower bound any divergence between the learned and target distributions. Second, we derive an upper bound on the Kullback-Leibler divergence that depends only on the observable gradient component of the error, tightening the standard Girsanov bound and identifying its looseness as the cost of operating on path-space rather than marginal-space dynamics. Third, we give a tractable estimator of the gradient component via a dual Sobolev identity, which is shown to empirically correlate substantially better with sample quality than the full $L^2$ error.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.06171unread
Effective Dimensionality as an Operator Invariant for Physics-Preserving Constraint Adaptation in Physics-Informed Neural Networks
Cornelius Otchere, Michael Shields · 2026-06-05
arXiv:2606. 06171v1 Announce Type: new Abstract: Physics-Informed Neural Networks inherently suffer from task interference because they rely on a shared parameter space to satisfy both governing differential equations and boundary conditions.
Read next because Effective Dimensionality as an Operator Invariant for Physics-Preserving Constraint Adaptation in Physics-Informed Neural Networks overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: class, rect, under, width, line, rate, project, without. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06171v1 Announce Type: new Abstract: Physics-Informed Neural Networks inherently suffer from task interference because they rely on a shared parameter space to satisfy both governing differential equations and boundary conditions. We analyze this structural conflict using the Fisher Information Matrix to quantify the effective degrees of freedom ($d_{eff}$) in a physics-constrained model. Unlike the classical $d_{eff}$ which measures how many parameter directions are informed by data against a statistical prior, our $d_{eff}$ measures the dimension of the parameter directions unconstrained by the differential operator. For operators with finite-dimensional kernel, we show that $d_{eff}$ converges to the kernel dimension exactly, independent of network width, depth, or activation function, recasting it from a fit diagnostic into a structural invariant of the underlying continuous operator. For operators with infinite-dimensional kernel, $d_{eff}$ instead measures the network's finite-dimensional representational bandwidth for that kernel rather than recovering an integer invariant. Importantly, $d_{eff}$ also serves as an a priori structural diagnostic. Driving $d_{eff}$ of a well-posed problem to zero certifies that the physics and boundary constraints have absorbed the network's free directions. Building on this characterization, we introduce subspace projection strategies for boundary adaptation. Rather than retraining from scratch, we project parameter updates into the null space of the pre-trained physics operator so that new boundary conditions are satisfied without disturbing the learned physics. Gradient-based fine-tuning can match or exceed this but needs more wall-clock time and tuning, whereas subspace projection delivers near-equivalent quality in seconds to minutes. We validate on linear and nonlinear operators, demonstrating accurate adaptation to initial and boundary shifts and unencountered constraint types.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05967unread
Fast and Robust Convergence Rate for TD(0) with Linear Function Approximation, Universal Learning Steps and I.I.D. Samples
Ziad Kobeissi (L2S), \'Elo\"ise Berthier (U2IS) · 2026-06-08
arXiv:2606. 05967v2 Announce Type: replace Abstract: In this paper, we study the finite-time behavior of the TD(0) temporal-difference method with linear function approximation (LFA).
Read next because Fast and Robust Convergence Rate for TD(0) with Linear Function Approximation, Universal Learning Steps and I.I.D. Samples overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, under, line, rate, does, chain, on-policy, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05967v2 Announce Type: replace Abstract: In this paper, we study the finite-time behavior of the TD(0) temporal-difference method with linear function approximation (LFA). We consider on-policy independent and identically distributed (i.i.d.) samples, a constant learning step, and the Polyak-Juditsky averaging method. We establish a new convergence rate, for the Mean-Square Error (MSE) on the approximated function, that is (i) fast in the sense that it admits an optimal dependency in the number of iterations k (i.e., of order 1/k), (ii) robust to ill-conditioning: it only depends on an initial error and modelindependent constants and (iii) sharp up to a multiplicative constant lower than 11. In particular, it does not depend on the smallest eigenvalue of the uncentered covariance matrix of the linear parametrization, unlike all pre-existing O(1/k) rates in the TD(0) literature. We also introduce PCTD(0), a variant of TD(0), which benefits from better convergence properties under an additional assumption of strong mixing on the Markov Chain.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05488unread
Sparse Functional Singular Value Decomposition for Biclustering and Triclustering Longitudinal Data
Yue Zhao, Thierry Chekouo, Sandra Safo · 2026-06-05
arXiv:2606. 05488v1 Announce Type: new Abstract: Identifying subtypes of complex conditions, such as Inflammatory Bowel Disease (IBD), often requires capturing latent patterns in longitudinal omics data.
Read next because Sparse Functional Singular Value Decomposition for Biclustering and Triclustering Longitudinal Data overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, rate, position. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05488v1 Announce Type: new Abstract: Identifying subtypes of complex conditions, such as Inflammatory Bowel Disease (IBD), often requires capturing latent patterns in longitudinal omics data. However, these data are typically high-dimensional, sparsely sampled, and irregularly observed over time, posing substantial challenges for conventional (bi)clustering and functional data analysis methods. We propose Tri-SfSVD, a unified sparse functional Singular Value Decomposition framework for discovering biclusters and triclusters in longitudinal data. Unlike existing functional biclustering methods that rely on ad hoc imputation or enforce restrictive shape-homogeneity assumptions, Tri-SfSVD integrates continuous trajectory estimation with simultaneous subject, feature, and temporal selection within a single optimization framework. By imposing sparse penalties across subjects, variables, and temporal subregions, the proposed method works directly on observed data to uncover localized structures at the subject, subject-feature, and subject-feature-time levels. Extensive simulations demonstrate that Tri-SfSVD outperforms existing approaches in high-dimensional settings. Applied to IBD multi-omics data, the method identified three biclusters linking sample clusters with distinct IBD-related clinical characteristics to microbial pathway groups associated with specific bacterial taxa, providing interpretable subject-pathway associations for characterizing disease heterogeneity. Applied to multi-channel EEG data, the method identified three triclusters linking sample clusters with distinct alcohol-related phenotypes to localized brain activity patterns, including subgroup differences separated by temporal subregions within the same spatial region.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05365unread
Environment-Robust Representation Learning with Empirical Bayes
Yuli Slavutsky, Matthew Shen, Bohan Wu, David M. Blei · 2026-06-05
arXiv:2606. 05365v1 Announce Type: new Abstract: We consider multi-environment prediction problems.
Read next because Environment-Robust Representation Learning with Empirical Bayes overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: eval, source, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05365v1 Announce Type: new Abstract: We consider multi-environment prediction problems. We assume the environments change the distribution of a latent variable, while the mechanisms generating observed covariates and targets remain stable conditional on that variable. For example, hospitals or clinical cohorts may differ in the prevalence of latent patient states, even though the relationships between those states, physiological measurements, and outcomes remain unchanged. Given a dataset from multiple environments, we formulate a Bayesian model for such problems and derive the corresponding variational objective. We show that this objective decomposes into per-environment terms and an additional cross-environment balancing term induced by the model's structure. We use an empirical Bayes method to set the prior and incorporate it into the objective. Based on this objective, we develop an amortized variational algorithm for posterior approximation, and use the resulting learned latent variables to form predictions in new environments.We study our approach through simulations and real-world studies of astronomical source identification, microbiome-based disease detection, and ICU sepsis prediction. Across these settings, our method outperforms previous approaches for prediction in new environments.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05361unread
TabSODA: Tabular Diffusion based Imputation with Skip Pattern Detection and Ordinal Awareness
Yuyu Chen, Taehyo Kim, Hai Shu, Yang Feng · 2026-06-05
arXiv:2606. 05361v1 Announce Type: new Abstract: Missing data imputation in large-scale surveys faces two challenges that are not well handled by current tabular diffusion methods.
Read next because TabSODA: Tabular Diffusion based Imputation with Skip Pattern Detection and Ordinal Awareness overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, rect, line, propagate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05361v1 Announce Type: new Abstract: Missing data imputation in large-scale surveys faces two challenges that are not well handled by current tabular diffusion methods. First, \emph{structural skips}, cells made inapplicable by questionnaire design, should not be imputed but are often conflated with item nonresponse. Second, \emph{ordinal} responses encode ordered categories, yet most pipelines treat them as nominal levels through one-hot or analog-bit encodings. We introduce \textbf{TabSODA} (\textbf{Tab}ular diffusion with \textbf{S}kip pattern detection and \textbf{O}r\textbf{d}inal \textbf{A}wareness), an Expectation-Maximization (EM)-based diffusion imputer built on the Elucidated Diffusion Model (EDM) framework. TabSODA propagates structural skips through the denoising loss and reverse-time sampler, and represents ordinal variables with cumulative-probit scalar latents while retaining analog-bit encodings for nominal variables. When a codebook skip mask is available, TabSODA uses it directly; otherwise, the TabSODA+SKIP variant estimates the mask from raw responses and questionnaire order using a CART-based skip-pattern miner. On Population Assessment of Tobacco and Health (PATH) study and the National Survey on Drug Use and Health (NSDUH), two nationally representative U.S.\ surveys, TabSODA reduces ordinal MACE by up to $23.7\%$ and improves categorical accuracy by up to $9\%$ over the strongest baseline across MCAR, MAR, and MNAR masking. The skip miner achieves near-perfect precision on both datasets, allowing TabSODA+SKIP to closely track the codebook-mask variant.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05945unread
Exploring the connection between coding habits and cognitive styles in malware developers
Vasilis Vouvoutsis, Constantinos Patsakis, Fran Casino · 2026-06-05
arXiv:2606. 05945v1 Announce Type: new Abstract: Malware research primarily studies the results, the methods, and the impact.
Read next because Exploring the connection between coding habits and cognitive styles in malware developers overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, class, soft, source, rate, compare, does, full. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05945v1 Announce Type: new Abstract: Malware research primarily studies the results, the methods, and the impact. Even from an offensive security perspective, what is examined is the method, not the development strategy of the offender. This study investigates the behavioral signatures and coding patterns embedded in the malware source code. By analyzing a large corpus of leaked malware code and comparing it with carefully selected benign open-source software, we apply static application security testing and compute multiple software metrics. Based on cognitive psychology and criminological theories, our work interprets differences in code structure and quality as behavioral indicators, reflecting distinct motivational structures, risk tolerances, and development strategies of malware authors compared to benign software developers. Our findings reveal that malware code is generally smaller, less documented, and exhibits higher cyclomatic complexity per function, with reduced use of abstraction mechanisms such as classes and closures. Vulnerability analysis further reveals that malware exhibits more issues of the types that benign code typically avoids, suggesting a minimal investment in secure development practices. These patterns imply a development style optimized for expedience, operational secrecy, and evasion rather than long-term maintainability. Nonetheless, the code quality metrics indicate that it does not deviate significantly from benign software enough to be distinctive. By framing code metrics as proxies for behavioral signals and strategic choices, we demonstrate how quantitative software analysis can enrich behavioral cybersecurity research, offering new insights into the practices and priorities of malware developers. Our results pave the way for further research in the behavioral profiling of cyber offenders.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05834unread
Towards Worst-case Hardness for Low-Noise LPN
Divesh Aggarwal, Rishav Gupta, Hai Hoang Nguyen, Kel Zin Tan, Prashant Nalini Vasudevan · 2026-06-05
arXiv:2606. 05834v1 Announce Type: new Abstract: The hardness of the Learning Parity with Noise (LPN) problem is a foundational assumption in cryptography, forming the basis of constructions ranging from symmetric-key primitives to public-key encryption and beyond.
Read next because Towards Worst-case Hardness for Low-Noise LPN overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, word, alpha, line, rate. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05834v1 Announce Type: new Abstract: The hardness of the Learning Parity with Noise (LPN) problem is a foundational assumption in cryptography, forming the basis of constructions ranging from symmetric-key primitives to public-key encryption and beyond. A central open question is whether the average-case hardness of LPN can be based on worst-case complexity assumptions, as has been achieved for the analogous Learning With Errors (LWE) problem. Existing worst-case-to-average-case reductions for LPN [BLVW19, YZ21] rely on statistical smoothing of linear codes, which inherently limits the resulting average-case hardness to noise rates as large as $1/2 - 1/\mathrm{poly}(n)$, which is insufficient for public-key applications. We explore a new approach towards obtaining such reductions: rather than requiring that random sparse combinations of the rows of the generator matrix of a code be statistically close to uniform, we only require that they be computationally indistinguishable from uniform. This leads to a clean win-win structure: we show that any efficient LPN solver can be transformed into a pair of efficient algorithms $(S, D)$ such that for every matrix $A$ of appropriate dimensions over $\mathbb{F}_2$, either $S$ decodes the code generated by $A$ from random noise, or $D$ distinguishes random noisy codewords of the dual of this code from uniform. By instantiating this reduction with appropriate parameters, we obtain the average-case hardness of LPN with inverse-polynomial noise rate $n^{-\alpha}$ for any constant $\alpha < 1$, assuming the worst-case simultaneous hardness of decoding a code from random noise and distinguishing random noisy codewords of its dual from uniform. In particular, setting $\alpha = 1/2$, our reduction yields LPN hardness in the parameter regime required for Alekhnovich's construction of public-key encryption [Ale03], a regime that was previously inaccessible via worst-case reductions.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05796unread
GCD: Garbled, Corrected, Demonstrandum -- Fixing and Proving Go's Extended GCD Implementation
Linard Arquint · 2026-06-05
arXiv:2606. 05796v1 Announce Type: new Abstract: We verify the 'extendedGCD' implementation in Go's standard library ('crypto/internal/fips140/bigmod'), which plays a crucial role in the generation of RSA key pairs.
Read next because GCD: Garbled, Corrected, Demonstrandum -- Fixing and Proving Go's Extended GCD Implementation overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, correct, line, implement. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05796v1 Announce Type: new Abstract: We verify the 'extendedGCD' implementation in Go's standard library ('crypto/internal/fips140/bigmod'), which plays a crucial role in the generation of RSA key pairs. Even though the Go implementation is supposedly a direct port from BoringSSL's implementation, we uncovered two deviations that each break the algorithm's invariants: (1) the Go implementation deviates in the way coefficients are updated, and (2) it permits a larger input domain. We address both deviations; the first by fixing the Go implementation, which results in an on average 24% speedup, and the second deviation by porting an existing proof for BoringSSL and extending it to cover the larger input domain. We prove correctness and termination of the fixed Go implementation using Gobra, a deductive program verifier for Go. Where necessary, we used Lean to prove key lemmata on non-linear arithmetic, which we import into Gobra. Our verification effort reveals three key insights: subtle bugs can slip into even well-reviewed code with surprising ease; formal verification is a powerful tool for uncovering them; and AI agents can facilitate the verification process by iteratively refining invariants and lemmata based on Gobra's error messages.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05776unread
An Improved CNN-LSTM Based Intrusion Detection System for IoT Networks
Mohammad Tariq Ikhlas, Pohanyar Khowaja Khil, Malik Muhammad Mueed Aslam, Muhammad Khuram Shahzad · 2026-06-05
arXiv:2606. 05776v1 Announce Type: new Abstract: With the rapid proliferation of IoT devices, security concerns have dramatically escalated and intrusion detection systems have become critical for protecting networked environments.
Read next because An Improved CNN-LSTM Based Intrusion Detection System for IoT Networks overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, eval, rate, capability, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05776v1 Announce Type: new Abstract: With the rapid proliferation of IoT devices, security concerns have dramatically escalated and intrusion detection systems have become critical for protecting networked environments. This paper presents an improved CNN-LSTM based intrusion detection model that combines multi-class classification, dataset integration, and temporal feature learning to enhance detection performance in IoT networks. Using network traffic data, the proposed approach is evaluated on intrusion detection tasks and achieves an accuracy of approximately 97%. Experimental results demonstrate that the model effectively detects multiple attack categories while maintaining stable training and validation performance. The integration of convolutional and recurrent neural network components enables the framework to capture both spatial and temporal characteristics of network traffic, improving overall intrusion detection capability in IoT environments.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05743unread
Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense
Minseok Choi, Seungbin Yang, Dongjin Kim, Subin Kim, Jungmin Son, Yunseung Lee, Jaegul Choo, Youngjun Kwak · 2026-06-05
arXiv:2606. 05743v1 Announce Type: new Abstract: Despite advances in safety alignment, large language models remain vulnerable to continuously evolving jailbreaks.
Read next because Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, class, under, alignment, rate, without, language, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05743v1 Announce Type: new Abstract: Despite advances in safety alignment, large language models remain vulnerable to continuously evolving jailbreaks. Existing fine-tuned safety classifiers cannot adapt to these evolving attacks, while adaptive memory-based guardrails tend to over-refuse benign queries that resemble stored attacks. We propose Membrane, a self-evolving guardrail built on Contrastive Safety Memory (CSM): each cell pairs the conditions for blocking a harmful query with those for permitting a superficially similar benign request. Without retraining, Membrane evolves CSM by distilling each harmful interaction and its benign counterpart into a contrastive cell indexed by the underlying attack strategy, so that one cell generalizes across topical variants of the same mechanism. At inference, retrieved cells serve as grounding context for precise safety decisions. Across model-level safety on HarmBench and agent-level safety on AgentHarm, Membrane achieves the highest F1 on all six jailbreak attacks. Notably, benign refusal on AgentHarm stays at 7-14%, well below the 28-85% range of prior guards. Memory cells also retain 87-88% F1 under cross-attack transfer and remain stable under memory poisoning.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05710unread
Explainable AI-Driven Cyber Risk Analytics and Model Reliability Assessment for Intelligent Governance of U.S. Critical Infrastructure: An XGBoost and SHAP-Based Intrusion Detection Framework
B. M. Taslimul Haque, Md. Arifur Rahman, Md. Serajul Kabir Chowdhury Rubel, Md. Iqbal Hossan · 2026-06-05
arXiv:2606. 05710v1 Announce Type: new Abstract: The increasing penetrations of the critical infrastructure sector in the United States with intelligent digital technologies have greatly increased exposure to advanced cyber adversaries and operational vulnerabilities.
Read next because Explainable AI-Driven Cyber Risk Analytics and Model Reliability Assessment for Intelligent Governance of U.S. Critical Infrastructure: An XGBoost and SHAP-Based Intrusion Detection Framework overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: class, rate, test, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05710v1 Announce Type: new Abstract: The increasing penetrations of the critical infrastructure sector in the United States with intelligent digital technologies have greatly increased exposure to advanced cyber adversaries and operational vulnerabilities. AI-powered governance and automated decision-making systems are becoming a key part of the operation of critical infrastructure systems, including energy, healthcare, transportation, financial services, and communication infrastructure, in order to improve efficiency and strategic management. The growing cyber threat environment, such as Distributed Denial of Service (DDos) attacks, botnets, ransomware, and Advanced Persistent Threats (APTs) pose significant challenges to infrastructure resilience, cyber security reliability, and governance trustworthiness. In a changing attack landscape and dynamic network environment, traditional cybersecurity mechanisms can often fall short of meeting the evolving needs and protecting critical systems. This study will develop a resilient cyber risk analytics and model reliability assessment framework to support intelligent governance and decision support for cyber risk exposure in the U.S. critical infrastructure environment. This study is based on the CICIDS2017 dataset for the development and testing of intrusion detection system models and cyber risk prediction models based on machine learning. Various classifiers like XGBoost, Random Forest, and Decision Tree are used to detect malicious activities on the network and determine the level of cyber risk. Furthermore, the Explainable Artificial Intelligence (XAI) techniques are integrated to enhance transparency, interpretability, and trust in cybersecurity decision-making processes. The proposed framework presents the reliability and resilience of the model by having various performance measures such as accuracy, precision, recall, F1 score, ROC-AUC, and false positive rate.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05648unread
Protecting K-Nearest Neighbor Queries from Location Inference Attacks
Zhiyu Sun, Jie Fu, Xinpeng Ling, Huifa Li, Zhili Chen · 2026-06-05
arXiv:2606. 05648v1 Announce Type: new Abstract: The k-nearest neighbor query (kNNQ) is a core component of modern location-based services (LBS) and has been widely adopted in popular features such as ``people nearby''.
Read next because Protecting K-Nearest Neighbor Queries from Location Inference Attacks overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rate, trained. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05648v1 Announce Type: new Abstract: The k-nearest neighbor query (kNNQ) is a core component of modern location-based services (LBS) and has been widely adopted in popular features such as ``people nearby''. However, its potential privacy risks have long been overlooked. In this work, we present the first two attacks against kNNQ, namely the geometric intersection location inference attack (GI-LIA) and the zero-order optimization location inference attack (ZO-LIA), revealing the inherent location privacy risks posed by kNNQ. To mitigate these privacy risks, we further propose DPRS, a differential privacy framework for kNNQ protection. The core idea of DPRS is to incorporate a rejection sampling mechanism within a constrained perturbation interval, thereby mitigating the distance distortion caused by excessive noise injection. In addition, we design a private interval construction algorithm to construct the perturbation interval, enabling the rejection sampling mechanism to achieve a more favorable trade-off between privacy protection and query utility in kNNQ. Extensive experiments on real-world spatial datasets demonstrate that DPRS outperforms existing methods in both privacy protection and query utility. Our code is available at https://github.com/reanatom/DPRS.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05476unread
SHIELDS: Automating OS Hardening with Iterative Multi-Agent Remediation
Andrew Hamara, Dwight Horne, Aldehir Rojas, Timothy Kurniawan, Sophie Lamothe, Vishal Suresh, Nicholas Turoci, Lawrence Wong · 2026-06-05
arXiv:2606. 05476v1 Announce Type: new Abstract: Security misconfigurations remain a leading cause of OS-level compromise, and manually keeping systems compliant with standards like Defense Information Systems Agency (DISA) Security Technical Implementation Guides (STIGs) is a tedious and expensive process.
Read next because SHIELDS: Automating OS Hardening with Iterative Multi-Agent Remediation overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, correct, eval, implement, full, language, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05476v1 Announce Type: new Abstract: Security misconfigurations remain a leading cause of OS-level compromise, and manually keeping systems compliant with standards like Defense Information Systems Agency (DISA) Security Technical Implementation Guides (STIGs) is a tedious and expensive process. Existing compliance automation tools can reduce some of this burden, but they depend on static, pre-written corrective actions. In this paper, we introduce SHIELDS, a multi-agent system that uses large language models (LLMs) to approach OS hardening as an iterative, feedback-driven process. Instead of applying fixed remediations, SHIELDS continuously proposes fixes and refines them based on feedback from target system execution and validation scans. We evaluate the system across multiple virtual machine configurations using six contemporary LLMs ranging from 20B to 400B parameters, and find that SHIELDS successfully remediates up to 73% of scan findings. Our results also suggest that success in this setting depends less on model size (parameter count) than on effective tool use and information gathering, paving a practical path toward reducing the burden of security compliance in environments where compute is limited or security and privacy needs drive local model use.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05459unread
CRESS: Quantifying Vulnerabilities of Attack Scenarios in Hardware Reverse Engineering
Alexander Hepp, Matthias Ludwig, Michaela Brunner, Johanna Baehr, Georg Sigl · 2026-06-05
arXiv:2606. 05459v1 Announce Type: new Abstract: The safety, security, and reliability of microelectronic systems depend on a trustworthy, secured supply chain and design flow.
Read next because CRESS: Quantifying Vulnerabilities of Attack Scenarios in Hardware Reverse Engineering overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: class, rate, chain. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05459v1 Announce Type: new Abstract: The safety, security, and reliability of microelectronic systems depend on a trustworthy, secured supply chain and design flow. Globally distributed supply chains or unintentional design weaknesses leave the door open for attacks on the hardware level. These scenarios encompass counterfeiting, hardware trojans, or on-device attacks. For these, hardware reverse engineering (RE) results play a pivotal role. The ongoing publication of new RE-involved attacks motivated the development of the common RE scoring system (CRESS). The system enables a general classification of RE-involved scenarios for a common, consistent rating. In this work, the originally qualitative system is extended to a quantitative system. We performed an extensive interview study with experts in the field. The interview results allowed us to derive weights that measure the severity of different RE-involved attack categories. The weights form an equation that quantifies scenarios, resulting in the severity-indicating CRESS score. The score enables the coherent rating of novel scenarios, renders them comparable, and supports the development of effective countermeasures. To showcase the effectiveness of the quantitative CRESS Score, six selected case studies are rated qualitatively and quantitatively. The CRESS Score proves to be significantly more expressive than the industry-standard Common Vulnerability Scoring System (CVSS).
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05396unread
Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration
Cristina Carleo, Pietro Liguori, Naghmeh Ivaki, Domenico Cotroneo · 2026-06-05
arXiv:2606. 05396v1 Announce Type: new Abstract: Producing a labeled vulnerable code at scale is a recurring obstacle for learning-based vulnerability detection: mined corpora carry substantial label noise, and existing LLM-based augmentation propagates these inaccuracies because it transforms vulnerable seeds rather than synthesising vulnerabilities from a specification.
Read next because Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, rect, eval, rate, project, propagate. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05396v1 Announce Type: new Abstract: Producing a labeled vulnerable code at scale is a recurring obstacle for learning-based vulnerability detection: mined corpora carry substantial label noise, and existing LLM-based augmentation propagates these inaccuracies because it transforms vulnerable seeds rather than synthesising vulnerabilities from a specification. A complementary route is to start from safe code and ask an instruction-tuned LLM to inject a specified CWE (which would shift the labeling burden from open-ended detection to bounded binary confirmation) but safety-aligned code LLMs systematically refuse such prompts. This paper is a preliminary feasibility study of abliteration, a low-rank weight edit that orthogonally projects out the refusal direction in the residual stream, as a tool to remove this barrier. We use Python and CWE-89 (SQL injection) as a case study, evaluating the Qwen2.5-Coder-Instruct family at 3B, 7B, and 14B parameters on safe samples drawn from PromSec and SafeCoder, replicated three times per condition. We find that (i) refusal on injection prompts is strongly size- and prompt-context-dependent: the 14B refuses 100% of prompts, the 7B refuses 73% of PromSec but only 5% of SafeCoder, whereas the 3B is essentially never blocked; (ii) abliteration reduces refusal to zero or near-zero across all sizes while leaving syntactic validity above 93%, supporting the view that, in this setting, refusal can be detached from measured code-generation capability; and (iii) the post-abliteration injection rate remains capacity-bound (88-97% on the 14B, 89-90% on the 7B, and 25-48% on the 3B) separating willingness, which abliteration unlocks, from capability, which scales with parameters. Vulnerability verdicts are produced by a three-tool detector ensemble (CodeQL, Semgrep, Bandit) followed by manual adjudication by two authors on detector-positive outputs.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05252unread
From Attack Simulation to SIEM Rule: Deterministic Detection-as-Code Synthesis with Probe-Level Traceability
Alexandre Cristov\~ao Maiorano · 2026-06-05
arXiv:2606. 05252v1 Announce Type: new Abstract: Security teams routinely simulate attacks against their own systems to check whether their monitoring would catch a real intruder.
Read next because From Attack Simulation to SIEM Rule: Deterministic Detection-as-Code Synthesis with Probe-Level Traceability overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, line, alone, emit. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05252v1 Announce Type: new Abstract: Security teams routinely simulate attacks against their own systems to check whether their monitoring would catch a real intruder. These Breach-and-Attack-Simulation (BAS) tools surface findings, but the security information and event management (SIEM) systems that watch production need detection rules -- and today a human bridges that gap by hand, reading each finding and writing the corresponding Sigma rule (a vendor-neutral detection format). We show this translation can be partially automated when probes are drawn from a locked corpus, so each finding carries a stable identifier back to the originating probe. We describe a deterministic synthesis function that maps each finding to a starter Sigma rule through a small template library (N=23, indexed by categories from the OWASP LLM and Web Top 10), with a back-reference to the originating finding and its MITRE ATT&CK technique. On two locked corpora (17-probe LLM, 23-probe Web), every bypassed-probe finding yields a starter rule, and all 17/17 emitted rules parse and convert to Splunk and Elasticsearch backends. Replayed through a live OpenSearch SIEM, the LLM rules fire on 30% of a held-out AdvBench subset and 14% of HarmBench at 7.7% false positives on a benign baseline; the Web side is validated structurally, not against a held-out attack set. The contribution is a verifiable, byte-stable path from BAS finding to operator-deployable starter rule, re-derivable from the published corpus and template library alone -- trading the breadth of LLM-generative methods for exact reproducibility and a typed traceback from any fired alert to the originating probe.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05513unread
EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts
Yiming Lu, Sihang Zeng, Zhengxu Tang, Max Lau, Fei Liu, Wei Jin · 2026-06-06
arXiv:2606. 05513v1 Announce Type: new Abstract: Epidemic LLM forecasters are usually trained and evaluated as static supervised models, whereas operational pandemic forecasting is a streaming process in which labels arrive after predictions and disease regimes shift over time.
Read next because EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, rate, compare, trained, leakage, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05513v1 Announce Type: new Abstract: Epidemic LLM forecasters are usually trained and evaluated as static supervised models, whereas operational pandemic forecasting is a streaming process in which labels arrive after predictions and disease regimes shift over time. We study this mismatch in weekly COVID-19 hospitalization trend forecasting across five variant regimes. We introduce EpiEvolve, a self-evolving agent that wraps an LLM forecaster trained on the warm-start period and keeps its weights fixed during streaming. EpiEvolve adapts by storing forecast outcomes in a hierarchical episodic memory, reflecting on delayed labels, retrieving cases relevant to the current regime, and distilling recurring errors into strategic rules. The resulting context lets the forecaster reuse its own past predictions and outcomes in later weeks while following a chronological protocol that prevents future leakage. On the streaming dataset, EpiEvolve reaches $0.629$ average accuracy, compared with $0.561$ for the static backbone and $0.325$ for the external CDC ensemble, and reduces recovery lag after regime shifts from $5$ to $2$ weeks. Ablations show that reflection, strategic memory, and regime-aware retrieval each contribute to the gains.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05461unread
Output Type Before Quality: A Standards-Derived XAI Admissibility Rubric for Autonomous-Driving Safety
Abhinaw Priyadershi, Mandar Pitale, Jelena Frtunikj, Maria Spence · 2026-06-06
arXiv:2606. 05461v1 Announce Type: new Abstract: Safety standards for ML-based autonomous driving specify the kind of evidence an assurance case must contain (directed cause-and-effect chains, quantified interventional effects, named root-cause variables), yet the XAI literature is organised by output type and technique family (saliency maps, feature attribution, counterfactuals, causal graphs, language traces).
Read next because Output Type Before Quality: A Standards-Derived XAI Admissibility Rubric for Autonomous-Driving Safety overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: class, rect, wrong, implement, chain, stage, test, language. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05461v1 Announce Type: new Abstract: Safety standards for ML-based autonomous driving specify the kind of evidence an assurance case must contain (directed cause-and-effect chains, quantified interventional effects, named root-cause variables), yet the XAI literature is organised by output type and technique family (saliency maps, feature attribution, counterfactuals, causal graphs, language traces). SHAP, the most-recommended ADS XAI method, returns a ranked feature list that no implementation effort can convert into a directed chain (Fig.1). We name this mismatch the evidence-type gap. From AMLAS, ISO 26262, ISO21448, ISO/PAS 8800 we derive 19 testable evidentiary criteria across 7 lifecycle stages with representative clause-cited derivations and score six XAI method classes structurally. Causal XAI emerges as structurally required to satisfy the derived criteria at three stages: hazard identification (+62% rubric gap), incident investigation (+50%), and data management (+50%); the verdict set is stable across thresholds T in (0%, 50%]$ and survives a worst-case single-cell flip down to T = 25%. At the remaining four stages, correlational or language-based methods are comparable or sufficient. The rubric identifies structural admissibility (necessary but not sufficient for compliance): an admissible method's specific output content may still be wrong, and validating that fidelity (the edges a fitted SCM produces, the cause a trace names) is the open assurance challenge. A single-VLA proof of concept on 1,996 real-world driving clips (79,840 rows, ten splits) is consistent with each method's observed output type matching its rubric prediction. XAI method selection for ADS safety assurance should be driven by lifecycle-stage evidence demand, not by method popularity.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05429unread
Minimizing the Hidden Cost of Scales: Graph-Guided Ultra-Low-Bit Quantization for Large Language Models
Rayyan Abdalla, Amir Hussein, Min Wu, Dinesh Manocha · 2026-06-06
arXiv:2606. 05429v1 Announce Type: new Abstract: Post-training quantization (PTQ) is critical for the efficient deployment of large language models (LLMs).
Read next because Minimizing the Hidden Cost of Scales: Graph-Guided Ultra-Low-Bit Quantization for Large Language Models overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, distributional, rate, implement, compare, position, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05429v1 Announce Type: new Abstract: Post-training quantization (PTQ) is critical for the efficient deployment of large language models (LLMs). Recent ultra-low-bit PTQ methods rely on rigid weight-saliency assumptions or position heuristics, introducing substantial hidden scaling overhead. We propose SAGE-PTQ (Saliency-Aware Graph-guided Efficient PTQ), a novel ultra-low-bit quantization framework for LLMs that minimizes hidden scaling cost. SAGE-PTQ separates salient and unsalient weights using distributional statistics, then models subsampled unsalient weights as a sparse graph to estimate the optimal number of groups per layer. SAGE-PTQ applies dual-mode quantization, assigning multi-bit precision to salient weights and binarizing unsalient weights. To reduce scaling overhead, SAGE-PTQ uses one per-channel scale for salient weights and one scalar per unsalient group. Finally, SAGE-PTQ implements adaptive saliency thresholding to select the optimal saliency ratio per matrix. SAGE-PTQ achieves 1.03 weight bits and only 0.004 scaling bits per matrix on average, outperforming state-of-the-art methods such as BiLLM and PB-LLM. On LLaMA-3-8B, SAGE-PTQ achieves 6.74 WikiText2 perplexity, compared to 55.8 for BiLLM, while using less than 50% of BiLLM's GPU memory. On LLaMA-2-70B, SAGE-PTQ provides 1.5x faster decoding on one NVIDIA L40 GPU, demonstrating practical inference efficiency.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05357unread
An interpretable and trustworthy AI framework for large-scale longitudinal structure-pain association studies using data from the Osteoarthritis Initiative (OAI)
Jincheng Yu, Haoyang Li, Yiwen Liu, Shen Liu, Rachel Yuanbao Chen, C. Kent Kwoh, Hongxu Ding, Xiaoxiao Sun · 2026-06-06
arXiv:2606. 05357v1 Announce Type: new Abstract: Purpose: To develop an interpretable and trustworthy AI framework that combines deep learning based MRI Osteoarthritis Knee Score (MOAKS) prediction with interpretable statistical modeling to study structure-pain relationships at scale using data from the Osteoarthritis Initiative (OAI).
Read next because An interpretable and trustworthy AI framework for large-scale longitudinal structure-pain association studies using data from the Osteoarthritis Initiative (OAI) overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: class, rect, rate, factor, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05357v1 Announce Type: new Abstract: Purpose: To develop an interpretable and trustworthy AI framework that combines deep learning based MRI Osteoarthritis Knee Score (MOAKS) prediction with interpretable statistical modeling to study structure-pain relationships at scale using data from the Osteoarthritis Initiative (OAI). Materials and Methods: We first developed a deep learning framework to predict MOAKS features directly from knee MRIs and incorporated conformal prediction to provide prediction uncertainty quantification. This uncertainty-aware strategy enables explicit filtering of model outputs, retaining only high-confidence MOAKS predictions at the knee level. Second, we applied a longitudinal latent class mixed model (LCMM) to examine associations between key structural abnormalities and four complementary knee pain measurements. Results: Among the three MRI-defined abnormalities (i.e., bone marrow lesions (BML), cartilage loss (CART), and meniscal extrusion (ME)), our framework substantially improved the Matthews correlation coefficient (MCC) and some other metrics. For example, MCC increased from 0.69 to 0.91 for BML, from 0.45 to 0.80 for CART, and from 0.59 to 0.89 for ME. Using these high-confidence predictions, we expanded the sample size to 2,175 knees for the LCMM analysis. Two distinct pain trajectories were identified (rapid and stable pain progression). The estimated odds ratios (95% CI) for the rapid progression group were 1.62 (1.12-2.35) for BML, 1.83 (1.24-2.70) for CART loss, and 2.50 (1.75-3.57) for ME. Conclusion: These results highlight the importance of these structural abnormalities as risk factors for pain and functional progression in osteoarthritis.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05332unread
GITCO: Gated Inference-Time Context Optimization in TSFMs
Manya Pandey, Dhruv Kumar, Murari Mandal, Saurabh Deshpande · 2026-06-06
arXiv:2606. 05332v1 Announce Type: new Abstract: Patch-based Time Series Foundation Models (TSFMs) suffer from context poisoning: structurally anomalous patches capture disproportionate attention and silently degrade zero-shot forecast quality.
Read next because GITCO: Gated Inference-Time Context Optimization in TSFMs overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, without, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05332v1 Announce Type: new Abstract: Patch-based Time Series Foundation Models (TSFMs) suffer from context poisoning: structurally anomalous patches capture disproportionate attention and silently degrade zero-shot forecast quality. We propose improving TSFM accuracy at inference time by optimizing the input context rather than modifying model weights. We present GITCO (Gated Inference-Time Context Optimization), a lightweight three-component framework: Gate, Router, and Critic that selectively identifies and suppresses harmful patches without any parameter updates. Evaluated on TimesFM 2.5 across 53 GIFT-Eval datasets under K-fold cross-validation, GITCO achieves an average +1.95% MASE reduction on TimesFM 2.5 while capturing 89.9% of the improvement upper bound. We introduce context sensitivity profiles as a new characterizable property of TSFMs: the mapping from time series meta-features to expected accuracy improvement under inference-time context intervention, shaped jointly by model architecture and the statistical structure of the data.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05304unread
What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems
Chen Huang, Yuhao Wu, Wenxuan Zhang · 2026-06-06
arXiv:2606. 05304v1 Announce Type: new Abstract: Multi-agent systems (MAS) built on large language models are typically organized around roles, pipelines, and turn schedules, while the content that agents pass to one another is often left as unconstrained natural language.
Read next because What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, token, line, rate, project, trained. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05304v1 Announce Type: new Abstract: Multi-agent systems (MAS) built on large language models are typically organized around roles, pipelines, and turn schedules, while the content that agents pass to one another is often left as unconstrained natural language. However, this free-form communication can rapidly inflate token usage, consume the shared context window, and ultimately affect both system performance and inference cost. We analyze five common inter-agent communication strategies across two MAS topologies, finding that no fixed strategy is universally optimal. Instead, effective inter-agent messages consistently preserve action-centered information needed by downstream agents. Building on this, we propose the PACT (Protocolized Action-state Communication and Transmission), which treats inter-agent communication as a public state-update problem and projects each raw agent output into a compact action-state record before it enters shared history. Across different MAS topologies, PACT consistently improves the performance-cost trade-off, achieving comparable or stronger task performance with substantially fewer tokens. The gains extend to production coding harnesses: PACT lifts OpenHands' resolve rate at -10% tokens-per-resolved, and is resolve-neutral on SWE-agent while halving input tokens. Our code is publicly available at https://github.com/iNLP-Lab/PACT.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05545unread
Multilingual Detection of Alzheimer's Disease from Speech: A Cross-Linguistic Transfer Learning Approach
Nadine Yasser Abdelhalim, Emmanuel Akinrintoyo, Nicole Salomons · 2026-06-05
arXiv:2606. 05545v1 Announce Type: new Abstract: The development of multilingual Alzheimer's Disease Dementia (AD) detection models presents significant challenges due to the resource-intensive and time-consuming nature of language-specific model training.
Read next because Multilingual Detection of Alzheimer's Disease from Speech: A Cross-Linguistic Transfer Learning Approach overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, class, source, screen, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05545v1 Announce Type: new Abstract: The development of multilingual Alzheimer's Disease Dementia (AD) detection models presents significant challenges due to the resource-intensive and time-consuming nature of language-specific model training. We propose a novel solution using cross-language training to detect AD in languages beyond those used for model training. This study investigates multilingual deep learning models for detecting AD across different languages and cognitive impairment levels. Using datasets in English, Chinese, Arabic, and Hindi, we developed transformer-based models for binary AD classification. Our approach achieved F1 scores of 82\% across all languages, demonstrating strong cross-linguistic generalization. The rapid inference time (0.5 seconds) supports potential real-time screening applications, while consistent performance across languages indicates feasibility for global deployment.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05180unread
From Scoring to Explanations: Evaluating SHAP and LLM Rationales for Rubric-based Teaching Quality Assessment
Ivo Bueno, Babette B\"uhler, Philipp Stark, Tim F\"utterer, Ulrich Trautwein, Dorottya Demszky, Heather Hill, Enkelejda Kasneci · 2026-06-05
arXiv:2606. 05180v1 Announce Type: new Abstract: Automated scoring models are increasingly used to assign rubric-based quality ratings to complex language performances, including classroom transcripts, yet they typically provide little insight into why a particular score is produced.
Read next because From Scoring to Explanations: Evaluating SHAP and LLM Rationales for Rubric-based Teaching Quality Assessment overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, eval, rate, trained, test, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05180v1 Announce Type: new Abstract: Automated scoring models are increasingly used to assign rubric-based quality ratings to complex language performances, including classroom transcripts, yet they typically provide little insight into why a particular score is produced. We propose a general framework for sentence-level interpretability of rubric-based scoring that combines model-agnostic Shapley-value attributions with rationales generated by large language models (LLMs). Instantiated on the Quality of Feedback dimension of the CLASS framework using the NCTE corpus, the framework enables systematic comparison of fine-tuned pretrained language models (PLMs) and prompted LLMs on both scoring performance and explanation faithfulness. Across 6k annotated transcript segments, fine-tuned PLMs outperform LLMs in prediction accuracy but exhibit label compression toward mid-scale scores. Deletion-based tests show that SHAP identifies sentences that reliably drive model predictions, producing typically larger and more coherent prediction shifts than LLM-generated rationales. Cross-model analyses further reveal that SHAP attributions transfer robustly across architectures, whereas LLM rationales exert limited and inconsistent influence. Overall, the findings demonstrate that SHAP provides more faithful and transferable explanations for rubric-based scoring, and that the proposed framework offers a principled basis for evaluating both scoring models and their explanations in high-stakes educational settings and other rubric-based language assessment tasks.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05174unread
Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO
Arash Ahmadi, Parisa Masnadi, Sarah Sharif, Charles Nicholson, David Ebert, Mike Banad · 2026-06-05
arXiv:2606. 05174v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong promise in healthcare applications.
Read next because Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, rate, full, on-policy, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05174v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong promise in healthcare applications. Yet deploying general-purpose models in real-world settings remains difficult due to data privacy constraints, inference costs, and limited suitability for edge or on-device use. These challenges motivate the development of smaller, more efficient models that require robust post-training strategies to ensure reliable medical reasoning. In this work, we investigate Group Relative Policy Optimization (GRPO) for post-training LLMs on heart-focused medical question answering with rubric-based supervision derived from RaR-Medicine. We propose a Variance-Aware Reward Framework that extends the Explicit Aggregation and Implicit Aggregation strategies of Rubrics as Rewards by replacing weighted binary criterion aggregation and single overall Likert-style scoring with continuous analytical reward functions derived from criterion-level rubric outcomes. This formulation provides richer optimization signals for feedback that is sparse, multi-criteria, and difficult to verify automatically, and enables more stable on-policy reinforcement learning. On a held-out heart-related subset of HealthBench, our best GRPO variant improves accuracy from 0.362 to 0.502 and F1 from 0.532 to 0.668 relative to the Qwen3-14B base model, while remaining competitive with GPT-OSS-120B (0.508 accuracy, 0.674 F1). Our findings show that carefully designed rubric-based rewards provide a practical strategy for improving heart-focused medical question answering in LLMs, with potential to extend to other rubric-based tasks.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05168unread
Epidemiology of Model Collapse: Modeling Synthetic Data Contamination via Bilayer SIR Dynamics
Xiangyu Wang · 2026-06-05
arXiv:2606. 05168v1 Announce Type: new Abstract: Training on synthetic data causes model collapse, but existing analyses treat this as single-chain degradation.
Read next because Epidemiology of Model Collapse: Modeling Synthetic Data Contamination via Bilayer SIR Dynamics overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, source, rate, chain, trained, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05168v1 Announce Type: new Abstract: Training on synthetic data causes model collapse, but existing analyses treat this as single-chain degradation. In reality, the AI ecosystem involves cross-contamination: models ingest synthetic data from other models, produce new synthetic text, and contaminate shared corpora. We propose a bilayer coupled SIR/SIRS framework -- a phenomenological mean-field model treating data corpora and AI models as two interacting populations, each with susceptible, infected, and recovered compartments linked by cross-layer transmission. The SIRS variant (our primary recommendation) incorporates immunity waning, reflecting that filtered corpora and retrained models remain susceptible to re-contamination. We derive the basic reproduction number $R_0 = \sqrt{\beta_D \beta_M / [(\gamma_D+\mu_D)(\gamma_M+\mu_M)]}$ via the Next Generation Matrix and apply standard epidemic threshold results to the bilayer system. Illustrative scenario-based calibration from public AI text prevalence data yields supercritical dynamics ($R_0 > 1$) across three scenarios; Sobol sensitivity analysis identifies synthetic-text detection as the highest-leverage parameter. A bipartite-network agent-based model confirms mean-field consistency ($R^2 > 0.96$) for dense networks but degrades under heterogeneity. GPT-2 contamination chain experiments (192 runs across WikiText and Shakespeare) show dose-response degradation and diversity loss qualitatively consistent with the threshold picture. Matched-budget source-diversity experiments (1,088 runs) provide suggestive evidence that multi-source mixing modestly attenuates collapse, but the effect vanishes at lower contamination fractions. Intervention analysis identifies detection-based filtering and herd immunity as the highest-leverage strategies.
source ↗open
score 98arxiv cs.LG (Machine Learning)arxiv:2606.04063unread
LLM Compression with Jointly Optimizing Architectural and Quantization choices
Hoang-Loc La, Truong-Thanh Le, Amir Taherkordi, Phuong Hoai Ha · 2026-06-05
arXiv:2606. 04063v1 Announce Type: new Abstract: Deploying large language models (LLMs) is challenging due to their significant memory and computational requirements.
Read next because LLM Compression with Jointly Optimizing Architectural and Quantization choices overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: line, rate, trained, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04063v1 Announce Type: new Abstract: Deploying large language models (LLMs) is challenging due to their significant memory and computational requirements. While some methods address this by developing small or tiny language models from scratch, these approaches demand extensive GPU training. Compressing pre-trained LLMs for edge devices offers a compelling alternative. Beyond pruning and quantization, Neural Architecture Search (NAS) enables effective compression, yet prior NAS approaches often limit the search space and decouple architecture from quantization. We introduce a differentiable NAS framework that explores the entire space and jointly optimizes architectural configurations alongside mixed-precision quantization for linear layers of LLMs. Experiments demonstrate superior accuracy-latency trade-offs: our models achieve up to 1.4x faster inference than sequential NAS-then-quantization baselines at comparable accuracy, or up to 6% higher average accuracy across seven reasoning tasks at equivalent latency.
source ↗open
score 98arxiv cs.CL (NLP)arxiv:2606.05444unread
Multilingual Coreference Resolution via Cycle-Consistent Machine Translation
Adriana-Valentina Costache, Eduard Poesina, Silviu-Florin Gheorghe, Paul Irofti, Radu Tudor Ionescu · 2026-06-05
arXiv:2606. 05444v1 Announce Type: new Abstract: Coreference resolution is a core NLP task, having a broad range of downstream applications, e.
Read next because Multilingual Coreference Resolution via Cycle-Consistent Machine Translation overlaps with clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: source, line, rate, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05444v1 Announce Type: new Abstract: Coreference resolution is a core NLP task, having a broad range of downstream applications, e.g.~machine translation, question answering, document summarization, etc. While the task is well-studied in English, comparatively less attention is dedicated to coreference resolution in other languages, especially low-resource ones. To mitigate this gap, we propose a novel coreference resolution pipeline that harnesses machine translation (MT) from English to a target low-resource language, to generate or expand training data. To automatically validate the quality of the translated samples, we back-translate the samples and assess the similarity with the original English samples via cosine similarity in the latent space of a BERT model. The resulting similarity scores are integrated into the loss function to weight training samples according to their MT cycle consistency. Extensive experiments on four low-resource languages show that our pipeline brings significant performance gains in coreference resolution. Moreover, our pipeline enables accurate coreference resolution in languages where no previous corpora were available.
source ↗open
score 94arxiv stat.ML (Machine Learning)arxiv:2606.06043unread
Adaptive Learning Rates with Surrogate Probability for Follow-the-Perturbed-Leader
Jongyeong Lee, Junya Honda, Shinji Ito, Chansoo Kim · 2026-06-05
arXiv:2606. 06043v1 Announce Type: new Abstract: Follow-the-regularized-leader framework has shown effectiveness and flexibility in online learning problems, where the choice of learning rates are known to be crucial.
Read next because Adaptive Learning Rates with Surrogate Probability for Follow-the-Perturbed-Leader overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone". Matching terms: alpha, line, rate, without. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06043v1 Announce Type: new Abstract: Follow-the-regularized-leader framework has shown effectiveness and flexibility in online learning problems, where the choice of learning rates are known to be crucial. Recently, adaptive learning rates defined in terms of the arm-selection probabilities, obtained by solving convex optimization, have achieved improved best-of-both-worlds (BOBW) guarantees in various bandit problems. In contrast, BOBW guarantees for its computationally efficient alternative, follow-the-perturbed-leader (FTPL), remain relatively limited since its optimization-free nature ironically makes the design of adaptive, probability-dependent learning rates non-trivial. To address this challenge, we propose an adaptive learning rate for FTPL by introducing surrogate probability functions that can be computed only from the available quantities, without requiring the exact probabilities. Based on these learning rates with surrogate functions, we provide the BOBW guarantee for FTPL with Pareto perturbations for any shape parameter $\alpha >1$, generalizing prior results restricted to specific choices of $\alpha=2$. We further show the BOBW guarantees for FTPL with adaptive learning rates in the bandit problem with expert advices. Our approach preserves the computational simplicity of FTPL while enabling probability-dependent adaptivity, and the surrogate-based methodology may be of independent interest in other algorithmic frameworks beyond FTPL and learning rate designs.
source ↗open
score 94arxiv stat.ML (Machine Learning)arxiv:2606.05919unread
Finding Most Influential Sets
Lucas D. Konrad, Nikolas Kuschnig · 2026-06-08
arXiv:2606. 05919v2 Announce Type: replace Abstract: Identifying most influential sets (MIS) - size-$k$ subsets whose removal maximally changes a target estimand - is typically infeasible because it requires searching over $\binom{n}{k}$ subsets.
Read next because Finding Most Influential Sets overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: under, line, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05919v2 Announce Type: replace Abstract: Identifying most influential sets (MIS) - size-$k$ subsets whose removal maximally changes a target estimand - is typically infeasible because it requires searching over $\binom{n}{k}$ subsets. For estimands with linear-fractional leave-set-out effects, we show that MIS selection reduces to a one-parameter sequence of top-$k$ problems. Dinkelbach's method yields an algorithm with $\mathcal{O}(n)$ cost per iteration and finite termination. For fixed residualized inputs, the algorithm returns a globally optimal set for the univariate ratio objective, including the oracle-residualized partial linear model. With estimated nuisance functions, uniform denominator and generated-score stability imply approximation to the first-order oracle orthogonal-score objective; exact set recovery follows under a separation condition. Simulations and applications show that the method recovers exact MIS that were previously computationally inaccessible.
source ↗open
score 94arxiv cs.AI (Artificial Intelligence)arxiv:2606.05433unread
Zero knowledge verification for frontier AI training is possible
Pierre Peign\'e, Ky Nguyen, Paul Wang · 2026-06-06
arXiv:2606. 05433v1 Announce Type: new Abstract: Frontier AI governance frameworks increasingly use cumulative training compute as the primary criterion for designating high-impact models, but enforcement rests on self-reporting because no technical verification primitive for training exists.
Read next because Zero knowledge verification for frontier AI training is possible overlaps with experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation", experiment "Test FR↔IT bystander-spill symmetry at multi-seed + 5 phrasings — pooled-rate vs per-phrasing asymmetry from #239 fact-check". Matching terms: without, candidate, test, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05433v1 Announce Type: new Abstract: Frontier AI governance frameworks increasingly use cumulative training compute as the primary criterion for designating high-impact models, but enforcement rests on self-reporting because no technical verification primitive for training exists. Any future international agreement on frontier AI faces the same problem at higher stakes: coordinated regulation of technologies with significant externalities has historically rested on technical verification, without which agreements are declaratory. Recent governance analyses judge zero-knowledge proofs a promising candidate but currently impractical at frontier scale [26, 4]. We argue the impracticality is paradigm-bound rather than fundamental, and propose a verification architecture for frontier dense pre-training combining a pre-committed training specification, inter-node network observations, and on-the-fly Merkle commitments of intermediate computation, verified through a zero-knowledge Virtual Machine (zkVM) with native BF16/FP32 precompiles. The proof checks the actual floating-point computation the GPU performed rather than a fixed-point approximation, and preserves model-architecture confidentiality through a private training specification. The protocol produces three proof types: a genesis proof at initialisation, in-training step proofs across the run, and ex-ante attestations enforcing policy-relevant claims as running invariants, turning the training record into a governance-enforceable artefact. We estimate a deployable proof of concept within approximately 36 months at single-digit-percent training-side overhead, against a six-to-ten-year cycle for verification-grade custom silicon. Thirteen open research and engineering problems are catalogued as a research agenda for external contribution
source ↗open
score 90arxiv cs.CR (Cryptography and Security)arxiv:2606.06013unread
Cheating in Multiplayer Online Games: a Dataset
Hugo Bertin, Marc Dacier, Y\'erom-David Bromberg · 2026-06-05
arXiv:2606. 06013v1 Announce Type: new Abstract: Cheating poses a significant threat to the Multiplayer Online Games (MOG) industry by degrading player satisfaction and undermining the fairness in competitive gaming.
Read next because Cheating in Multiplayer Online Games: a Dataset overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: class, under, line. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.06013v1 Announce Type: new Abstract: Cheating poses a significant threat to the Multiplayer Online Games (MOG) industry by degrading player satisfaction and undermining the fairness in competitive gaming. Despite efforts to develop mitigation techniques, cheating remains difficult to detect and prevent in practice. In particular, a class of cheats based on network flow disruption remains unsolvable. To find out how to detect such attacks we need access to representative labelled data. However, no such dataset exists. To address this gap, we leverage an experimental framework that combines a multiplayer online game with a plug-in capable of both reproducing cheating attacks and collecting logs at two levels: network and application-layer. This paper presents a dataset compiling records of game sessions played by both real players and automated game clients, with cheating actions explicitly logged. To the best of our knowledge, this is the first dataset that provides logs of network flow disruption cheats. While it includes such network-based cheats, it is not limited to them and also contains records of more commonly studied cheats, such as aimbots and wallhacks. This dataset can be used by researchers in academia and industry seeking to develop cheating detection mechanisms for online games. Furthermore, it is designed to be evolutive and can be enriched by others creating their own data traces with the proposed framework.
source ↗open
score 86arxiv stat.ML (Machine Learning)arxiv:2606.06391unread
Conformal Risk Sharing: Certified Cost Allocation with Participation Guarantees
Ieva Kazlauskaite · 2026-06-05
arXiv:2606. 06391v1 Announce Type: new Abstract: Sharing the financial impact of rare adverse events across a group can soften extreme individual burdens, but any participant made worse off by the arrangement has reason to leave.
Read next because Conformal Risk Sharing: Certified Cost Allocation with Participation Guarantees overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone". Matching terms: under, soft, distributional, control, without. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06391v1 Announce Type: new Abstract: Sharing the financial impact of rare adverse events across a group can soften extreme individual burdens, but any participant made worse off by the arrangement has reason to leave. A credible mechanism must therefore provide each agent with a trustworthy cap on their future obligation and should be deployed only if the aggregate harm across participants is bounded. We formalise this as the Certified Allocation Problem: from finite data and without distributional assumptions, find a redistribution rule, produce obligation caps for every participant, and verify that no participant is made materially worse off. We propose Conformal Risk Sharing, which solves this problem by pairing an interpretable sharing policy with split conformal calibration. The sharing intensity is tuned on training data, while held-out calibration data produces distribution-free per-agent guarantees (valid under exchangeability). Experiments on synthetic and real-world data, including precipitation and energy-cooperative data, confirm that the framework can substantially reduce extreme obligations for high-risk agents while controlling harm to others.
source ↗open
score 82arxiv cs.AI (Artificial Intelligence)arxiv:2606.05411unread
A Motivational Architecture for Conversational AGI
Anna Mikeda, Ben Goertzel · 2026-06-06
arXiv:2606. 05411v1 Announce Type: new Abstract: Motivational architectures in cognitive AI have largely been designed for physical agents regulating bodily needs.
Read next because A Motivational Architecture for Conversational AGI overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: latin, line, rate, stage. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05411v1 Announce Type: new Abstract: Motivational architectures in cognitive AI have largely been designed for physical agents regulating bodily needs. Conversational agents operate in a different regime: their sensorimotor loop is linguistic, their environment is a user's evolving mental state, and their consequential actions are speech acts, tool invocations, and strategic silences. This paper proposes a conversational reinterpretation of the OpenPsi motivational lineage, coupled to MetaMo's higher-level motivational scaffold, for agents built on a modular execution substrate. Homeostasis is recast in dialogue-native terms: the agent regulates competence, uncertainty reduction, affiliation, affinity, legitimacy, nurturing, and aesthetic coherence rather than bodily deficits. We propose three contributions: a ten-stage motivational processing pipeline that architecturally separates cognitive modulation from situational appraisal; a dual decision strategy blending urgency-driven fast response with deliberative multi-goal optimization; and an architecturally useful distinction between pre-action feelings and post-action emotions as functionally different forms of affect. We specialize the framework to two example agents -- CompanionAgent and ResearchAgent -- and sketch its extension to social robotics and domain-generic human-level AGI.
source ↗open
score 78arxiv stat.ML (Machine Learning)arxiv:2606.06288unread
Discrete Causal Representations from Heterogeneous Domains: A Bayesian Approach with Social Survey Applications
Ankur Garg, Michael Stettler, Aaron Schein, Julius von K\"ugelgen · 2026-06-05
arXiv:2606. 06288v1 Announce Type: new Abstract: Causal representation learning aims to infer the high-level latent causal concepts that give rise to observed low-level measurements.
Read next because Discrete Causal Representations from Heterogeneous Domains: A Bayesian Approach with Social Survey Applications overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: under, soft, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06288v1 Announce Type: new Abstract: Causal representation learning aims to infer the high-level latent causal concepts that give rise to observed low-level measurements. This is particularly relevant for heterogeneous data from different environments or domains since distribution shifts often arise through sparse, localized changes in some of the underlying causal mechanisms, while other parts of the generative process remain unchanged. Whereas identifiability of causal representations has been studied extensively, practical uncertainty-aware methods and real-world use cases remain less explored. In this work, we propose a Bayesian approach to learning causal representations from multi-environment data, focusing on the case of discrete causal concepts and unknown multi-node soft interventions. To this end, we translate causal assumptions and interpretability desiderata into suitable priors and parametric choices within a hierarchical model. We then devise an inference scheme based on sequential Monte Carlo sampling to approximate the resulting multimodal posterior. We showcase our approach through case studies on social survey data, where latent causal concepts correspond to cultural values or political opinions, measurements to survey responses, and environments to different countries or states. Our model infers meaningful high-level concepts and plausible causal relations among them, demonstrating its utility for learning causal representations of complex real-world data.
source ↗open
score 78arxiv cs.AI (Artificial Intelligence)arxiv:2606.05464unread
Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces
Nicol\'as Astorga, Nabeel Seedat, Mihaela van der Schaar · 2026-06-06
arXiv:2606. 05464v1 Announce Type: new Abstract: Verifiable reward training has improved mathematical and coding reasoning, but these domains capture only part of step-by-step decision making.
Read next because Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone". Matching terms: eval, line, without. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05464v1 Announce Type: new Abstract: Verifiable reward training has improved mathematical and coding reasoning, but these domains capture only part of step-by-step decision making. Many real-world tasks require finding a high-value feasible plan among many valid alternatives. We introduce OPT*, a scalable family of optimization-style tasks for training and evaluating LLM step-by-step optimization-like reasoning along a complexity axis: each task provides a feasibility checker and evaluator, while a complexity parameter expands the search space without requiring new human labels. This motivates studying these tasks in two regimes: (i) solver-guided online policy optimization, which uses a solver as a value oracle for partial states and applies rank-based reward shaping to reinforce better next steps, and (ii) search-based offline RL when such solvers are unavailable. Theoretically, we relate success in large search spaces to the information a reasoner extracts per unit of search budget. Empirically, we ablate the ingredients that make search efficient on OPT* and show that training on OPT* improves step-by-step optimization-like reasoning.
source ↗open
score 62arxiv stat.ML (Machine Learning)arxiv:2606.05371unread
Mamba-Assisted Non-Markovian Closure for Reduced-Order Modeling
Zhi-Feng Wei, Saad Qadeer, Panos Stinis · 2026-06-05
arXiv:2606. 05371v1 Announce Type: cross Abstract: Reduced-order modeling of high-dimensional dynamical systems is often hindered by the non-Markovian closure term that represents the effect of unresolved variables on the resolved dynamics.
Read next because Mamba-Assisted Non-Markovian Closure for Reduced-Order Modeling overlaps with experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?", experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: trained, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05371v1 Announce Type: cross Abstract: Reduced-order modeling of high-dimensional dynamical systems is often hindered by the non-Markovian closure term that represents the effect of unresolved variables on the resolved dynamics. Inspired by the Mori--Zwanzig formalism, in which the closure takes the form of a memory functional of the resolved trajectory, we recast closure modeling as a sequence modeling problem and propose the Mamba-Assisted Closure (MAC) framework: a Mamba-based sequence model, trained to predict the closure from the resolved trajectory, is coupled with the reduced-order governing equations through a numerical integrator to advance the resolved variables in time. A key feature of the framework is its exploitation of the dual representation of state-space models -- the model is trained in a sequence-to-sequence fashion via the convolutional form, and deployed for step-by-step autoregressive rollout via the recurrent form, yielding both efficient long-trajectory training and constant per-step inference cost. On the viscous Burgers' equation and the chaotic two-scale Lorenz '96 system, the MAC model substantially outperforms the Markovian reduced-order model, the GRU-based sequence model, and the Wilks method in predictive accuracy and long-time rollout stability.
source ↗open
score 62arxiv stat.ML (Machine Learning)arxiv:2606.05551unread
Conformal Risk-Averse Decision Making with Action Conditional Guarantee
Zihan Zhu, Shayan Kiyani, George Pappas. Hamed Hassani · 2026-06-05
arXiv:2606. 05551v1 Announce Type: new Abstract: Reliable decision making pipelines powered by machine learning models require uncertainty quantification (UQ) methods that come with explicit safety guarantees.
Read next because Conformal Risk-Averse Decision Making with Action Conditional Guarantee overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: line, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05551v1 Announce Type: new Abstract: Reliable decision making pipelines powered by machine learning models require uncertainty quantification (UQ) methods that come with explicit safety guarantees. Conformal prediction provides such UQ by wrapping ML predictions into prediction sets, and recent work by Kiyani et al. (2025b) established that these sets can be translated into optimal risk-averse decision policies -- yet only inheriting marginal safety guarantees. We generalize and strengthen their results by (i) introducing action-conditional conformal prediction, which yields safety guarantees conditioned explicitly on each action taken by the decision maker, (ii) showing that action-conditional prediction sets serve as a proxy for the feasible decision space for risk-averse decision makers aiming to optimize action-conditional value-at-risk, and (iii) proposing a principled finite-sample algorithm based on pinball-loss minimization, connecting the framework of Gibbs et al. (2025) to action-conditional guarantees. Experiments on two real-world datasets confirm that our approach significantly improves action-conditional performance over conformal baselines.
source ↗open
score 62arxiv cs.AI (Artificial Intelligence)arxiv:2606.05420unread
Assessing the Carbon Emissions and Energy Consumption of U.S. Hyperscale Data Centers
Gianluca Guidi, Francesca Dominici, Tiziano Squartini, Callaway Sprinkle, Jonathan Gilmour, Kevin Butler, Eric Bell, Scott Delaney, Falco J. Bargagli-Stoffi · 2026-06-06
arXiv:2606. 05420v1 Announce Type: new Abstract: The rapid proliferation of hyperscale data centers (HDCs) in the US, mainly driven by the adoption of artificial intelligence, has raised concerns about this industry's environmental footprint.
Read next because Assessing the Carbon Emissions and Energy Consumption of U.S. Hyperscale Data Centers overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: under, source. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05420v1 Announce Type: new Abstract: The rapid proliferation of hyperscale data centers (HDCs) in the US, mainly driven by the adoption of artificial intelligence, has raised concerns about this industry's environmental footprint. We compiled facility-level information on 403 US hyperscale data centers operating between May 2024 and April 2025 and estimated their electricity consumption, electricity sources, and attributable CO2 emissions. Across different facility-load scenarios, these HDCs consumed approximately 68-99 TWh of electricity and were associated with about 37-54 million metric tons of CO2. Under the central scenario, HDC electricity demand corresponded to approximately 1.8% of total US electricity consumption, with roughly 54% of attributed generation supplied by fossil-fuel sources. The HDC electricity-weighted average carbon intensity was approximately 545 gCO2/kWh, about 48% above the contemporaneous US national grid-average carbon intensity of 370 gCO2/kWh. Our approach provides an attributional tool for assessing the environmental footprint of hyperscale data centers using the most recent EPA eGRID plant-level data.
source ↗open

New research

score 30arxiv cs.AI (Artificial Intelligence)arxiv:2606.05528unread
When Should We Protect AI? A Precautionary Framework for Consciousness Uncertainty
Anna Mikeda · 2026-06-06
arXiv:2606. 05528v1 Announce Type: new Abstract: Existing frameworks assess whether AI systems might be conscious but provide no guidance on what to do with that assessment.
Background read from arxiv cs.AI (Artificial Intelligence). It did not strongly match recent Sagan clean results, beliefs, or experiments, so keep it lower priority unless the title is independently relevant.
arXiv:2606.05528v1 Announce Type: new Abstract: Existing frameworks assess whether AI systems might be conscious but provide no guidance on what to do with that assessment. We address this gap with a precautionary framework that maps consciousness evidence to graduated protective obligations. The framework comprises three components: (1) five welfare-relevant dimensions--phenomenal consciousness, affective valence, metacognitive awareness, self-narrative, and agency--each grounded in established consciousness science and linked to distinct moral concerns; (2) a threshold-plus-gradation hybrid specifying both binary triggers for new obligation categories and continuous scaling of protective weight; and (3) two complementary approaches to cross-dimensional aggregation, one hierarchical (drawing on Bach and Sorensen's Machine Consciousness Hypothesis) and one architecture-agnostic. We operationalize the framework through worked case studies of Replika and OpenClaw, demonstrating how systems occupying different regions of the dimensional space trigger different obligations, and derive design guidance for developers building systems near consciousness-relevant thresholds. The framework is architecture-agnostic, applying across neural, symbolic, and neurosymbolic systems, and aims to make consciousness science decision-relevant for organizations navigating uncertainty today.
source ↗open

Threats and caveats

score 100arxiv cs.LG (Machine Learning)arxiv:2606.04180unread
KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models
Youqi Wu, Mohammad Jalali, Farzan Farnia · 2026-06-05
arXiv:2606. 04180v1 Announce Type: new Abstract: Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems.
Read next because KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, rect, under, alignment, eval, compare, project. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04180v1 Announce Type: new Abstract: Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems. While these models are typically compared through downstream performance, such evaluations often do not explain how their representations differ structurally. In this work, we study this problem through the task of Contrastive Embedding Clustering: identifying sample subsets that are weakly clustered under one representation but strongly clustered under another. We propose \emph{Kernel Optimization for Discrepancy Analysis (KODA)}, a kernel-based framework for contrastive representation comparison and alignment. KODA constructs unified multimodal kernels through modality-wise kernel composition and formulates discrepancy discovery as a constrained optimization problem that searches for coherent structures in one representation while suppressing coherence in a reference representation. This yields interpretable discrepancy directions associated with specific sample subsets and modality interactions. To scale KODA to large vision-language datasets, we develop randomized low-dimensional approximations of joint kernels using random projections, including Random Fourier Features for shift-invariant kernels. Empirically, KODA identifies consistent and interpretable discrepancy structures across vision-language representations and provides sample subsets for representation alignment. The code is available at https://github.com/yokiwuuu/KODA.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04168unread
When Autoregressive Consistency Hurts Safety Alignment
Bochen Lyu, Yiyang Jia, Xiaohao Cai, Zhanxing Zhu · 2026-06-05
arXiv:2606. 04168v1 Announce Type: new Abstract: Safety alignment in large language models (LLMs) is fragile in part because it is often shallow: fine-tuning mainly reshapes the model's behavior near the first few output tokens.
Read next because When Autoregressive Consistency Hurts Safety Alignment overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: class, rect, under, alignment, prefix, token, rate, position. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04168v1 Announce Type: new Abstract: Safety alignment in large language models (LLMs) is fragile in part because it is often shallow: fine-tuning mainly reshapes the model's behavior near the first few output tokens. We argue that this phenomenon can be understood through autoregressive consistency, the tendency of next-token prediction to preserve and extend the current response trajectory consistently. By analyzing the learning dynamics of safety alignment, we show that autoregressive consistency can concentrate alignment updates on early tokens, offering a mechanistic explanation for shallow safety alignment. The same mechanism also predicts a broader class of attacks on LLMs: attacks that induce harmful continuation states at arbitrary positions in the output trajectory. As a concrete example, we introduce random insertion attack, which inserts a short harmful span into an otherwise safe refusal trajectory and exploits autoregressive consistency to sustain the resulting harmful branch, thereby bypassing safety alignment. Notably, a short harmful span can redirect the generation to be harmful even after a long refusal prefix, highlighting autoregressive consistency as a potential broader failure mechanism. This suggests that safety alignment should also break harmful autoregressive consistency throughout the output trajectory. We therefore propose adversarial safety alignment, an initial framework based on worst-case harmful continuation states, and instantiate it with random worst-insertion training. Overall, our results suggest that autoregressive consistency should be treated as a central consideration in both safety alignment and attack design.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses failure, adversarial.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04164unread
ADAPTOOD: Uncertainty-Aware Fine-Tuning for Out-of-Distribution ECG Time Series Models
Sotirios Vavaroutas, Yu Yvonne Wu, Ali Etemad, Cecilia Mascolo · 2026-06-05
arXiv:2606. 04164v1 Announce Type: new Abstract: Data samples used for training often differ from those encountered during fine-tuning and deployment, and while ML models show promise, their performance remains limited when only small annotated datasets are available.
Read next because ADAPTOOD: Uncertainty-Aware Fine-Tuning for Out-of-Distribution ECG Time Series Models overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, rect, under, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04164v1 Announce Type: new Abstract: Data samples used for training often differ from those encountered during fine-tuning and deployment, and while ML models show promise, their performance remains limited when only small annotated datasets are available. Performance often degrades under distribution shifts caused by diverse sensors, populations, and application settings. Although pre-training helps, models frequently encounter out-of-distribution (OOD) data in real-world settings, leading to reduced robustness. Existing adaptation methods usually assume fixed distribution shifts and struggle when multiple types or severities occur. In particular, they overlook shift severity, for example treating adaptation to a large familiar dataset the same as adaptation to a small dataset with a new task, which limits generalisation. To address this, we propose ADAPTOOD, a novel framework that leverages data uncertainty to quantify distribution shift severity and guide fine-tuning for time series. This uncertainty measures how strongly samples from the target deployment distribution deviate from the pre-training distribution, providing a direct signal of OOD severity. Our framework combines this uncertainty with low-rank model updates and adaptive hyperparameter optimisation to improve adaptation. We show that ADAPTOOD achieves up to 7% higher accuracy and 12.9% higher precision than existing methods in OOD tasks, maintaining strong performance as distribution shift severity increases.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04143unread
Physics-Informed Machine Learning for Short-Term Flood Prediction
Tewodros Syum Gebre, Jagrati Talreja, Leila Hashemi-Beni · 2026-06-05
arXiv:2606. 04143v1 Announce Type: new Abstract: Accurate flood forecasting is essential for mitigating disaster risks and protecting communities.
Read next because Physics-Informed Machine Learning for Short-Term Flood Prediction overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: latin, rect, under, alignment, line, rate, without, trained. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04143v1 Announce Type: new Abstract: Accurate flood forecasting is essential for mitigating disaster risks and protecting communities. However, purely data-driven machine learning models often struggle in data-scarce environments and may violate fundamental hydrological principles. Standard Long Short-Term Memory (LSTM) networks can generate physically inconsistent predictions, particularly when extrapolating to extreme weather conditions. To address these limitations, we propose a Physics-Informed Machine Learning (PIML) framework that incorporates hydrological knowledge directly into the loss function of an LSTM model. Specifically, a Trend Alignment constraint penalizes directional inconsistencies between precipitation and discharge trends, improving model robustness without requiring complex hydrodynamic equations. This regularization encourages the model to learn physically plausible hydrograph behavior, even with limited training data, while enhancing reliability during peak flood events. Experimental results show that the proposed physics-informed model outperforms a standard LSTM baseline in data-scarce settings, increasing the Nash-Sutcliffe Efficiency (NSE) from 0.20 to 0.23 when trained on only 5% of the available data. Additional stress tests under simulated extreme climate scenarios demonstrate that the baseline model exhibits unstable behavior, whereas the physics-informed model maintains directional consistency and physical plausibility. Although accurately predicting extreme peak magnitudes remains challenging with limited data, the proposed approach substantially reduces unphysical fluctuations common in purely data-driven models. These findings demonstrate that simple physical constraints can significantly improve the reliability of deep learning models for real-time flood forecasting, offering a practical solution for ungauged basins and evolving climate conditions.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses limitation, limitations, robustness.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04135unread
Stationarity-Aware Retrieval-Augmented Time Series Forecasting
Shiqiao Zhou, Holger Sch\"oner, Zipeng Wu, Edouard Fouch\'e, IAG Wilson, Shuo Wang · 2026-06-05
arXiv:2606. 04135v1 Announce Type: new Abstract: Time series forecasting relies on historical patterns, but real-world series often exhibit non-stationarity and regime shifts that challenge fully parametric forecasters.
Read next because Stationarity-Aware Retrieval-Augmented Time Series Forecasting overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, under, eval, line, rate, does, full. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04135v1 Announce Type: new Abstract: Time series forecasting relies on historical patterns, but real-world series often exhibit non-stationarity and regime shifts that challenge fully parametric forecasters. Inspired by Retrieval-Augmented Generation (RAG), recent work augments forecasters by retrieving relevant historical segments and using them as external evidence at inference time. However, due to the intrinsic non-stationarity of real-world time series, a highly similar past segment does not necessarily imply a similar future, rendering similarity-only retrieval brittle and prone to redundancy. We propose Stationarity-Aware Retrieval-Augmented Time Series Forecasting (SARAF), a framework that adaptively balances relevance and diversity in retrieval. SARAF first forms a candidate pool via temporal similarity with time-aligned enhancement, then applies a diversity-aware selection strategy to cover heterogeneous historical regimes, with the diversification strength automatically modulated by dataset-level stationarity. Moreover, SARAF uses stationarity-aware aggregation to fuse the retrieved futures. Extensive experiments on eight real-world datasets show that SARAF achieves competitive forecasting performance and improves average accuracy and robustness over strong baselines, with particularly clear benefits under challenging non-stationary settings. Code: https://github.com/ShiqiaoZhou/SARAF.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04115unread
dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formats
Giuseppe Franco, Ian Colbert, Pablo Monteagudo-Lago, Felix Marty, Nicholas Fraser · 2026-06-05
arXiv:2606. 04115v1 Announce Type: new Abstract: Quantizing large language models (LLMs) to low-precision floating-point representations is central to efficient deployment, yet applying a single bit-width uniformly across all layers is sub-optimal in terms of both performance and accuracy.
Read next because dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formats overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, width, eval, project, without, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04115v1 Announce Type: new Abstract: Quantizing large language models (LLMs) to low-precision floating-point representations is central to efficient deployment, yet applying a single bit-width uniformly across all layers is sub-optimal in terms of both performance and accuracy. This work introduces dMX, a differentiable mixed-precision quantization framework for learnable floating-point bit-width assignment. We study its application for the microscaling floating-point (MXFP) family of data types defined by the Open Compute Project (OCP) standard. The per-layer bit-width assignment is formulated as a continuous optimization problem in which each layer's floating-point format format is parameterized by a scalar parameter, folding the multi-variate design space into a single learnable offset. During training this offset takes continuous values, avoiding sudden oscillations between discrete quantization formats. A temperature-based annealing schedule progressively discretizes the learned offsets, ensuring that the final configuration maps to hardware-compatible MXFP formats without abrupt transitions between training and inference behavior. A target-aware regularization term steers the average bit-width toward a user-specified budget, serving as a coarse-grained proxy for inference cost and balancing model quality against deployment efficiency. We performed experiments on different families of LLM, such as Llama, Qwen3, and SmolLM2, evaluating perplexity on WikiText-2 and accuracy on four zero-shot reasoning benchmarks. Across these settings, dMX consistently yields Pareto-dominating models and improves over Kullback-Leibler (KL) divergence-based layer-selection heuristics, efficiently navigating trade-offs between model quality and average bit-width.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04110unread
Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification
Neeti Pokharna, Olivier Jeunen, Yatharth Saraf, Aleksei Ustimenko · 2026-06-05
arXiv:2606. 04110v1 Announce Type: new Abstract: Online evaluation of ranking and retrieval systems often relies on downstream monetization metrics such as app revenue or creator earnings.
Read next because Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: under, eval, line, without. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04110v1 Announce Type: new Abstract: Online evaluation of ranking and retrieval systems often relies on downstream monetization metrics such as app revenue or creator earnings. These metrics are typically heavy-tailed, with a small fraction of users dominating both mean and variance, leading to low statistical power and unreliable conclusions in A/B experiments -- especially under limited traffic. We present a practical framework for variance reduction in online experiments by combining post-stratification with CUPED. Our approach leverages pre-experiment covariates to improve the sensitivity of monetization experiments without requiring additional traffic. Deployed at ShareChat across ranking-driven monetization experiments, the method substantially reduces variance and improves decision stability, achieving equivalent statistical confidence with ~45\% less traffic than standard metrics. We further discuss practical design choices, guardrails, and limitations, providing guidance on when post-stratification is appropriate for real-world information retrieval and Recommendation systems.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses limitation, limitations, evaluation.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04106unread
Building The Ph(ysical)AI Layer Of Machine Intelligence
Ulbert Jose Botero, Liam Smith, Brooks Olney, Pooya Khorrami, Steven Kusiak, Watson Jia, Sage Trudeau, Daniel Capecci · 2026-06-05
arXiv:2606. 04106v1 Announce Type: new Abstract: Foundation models achieve generalization through massive-scale training on diverse data, but have limitations with transfer to truly unseen domains without paired training data.
Read next because Building The Ph(ysical)AI Layer Of Machine Intelligence overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, text, under, line, without, position, symmetry, language. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04106v1 Announce Type: new Abstract: Foundation models achieve generalization through massive-scale training on diverse data, but have limitations with transfer to truly unseen domains without paired training data. We propose principle-driven foundation models that encode signal-theoretic principles (Fourier decomposition, energy conservation, symmetry) rather than learn untethered statistical correlations. We hypothesize that domains differ not in fundamental physics, but in learnable transformations in time, frequency, magnitude, or phase. Training exclusively on radio-frequency (RF) data with co-designed architecture and losses incorporating these principles, we achieve cross-modal transfer to audio, images, text, and video using only frozen representations learned from RF data, requiring no fine-tuning of the encoder on target domains. Our 1.99M parameter frozen encoder achieves 77.7% average accuracy (91.9% top-3) across 15 diverse tasks via linear probing, with systematic variation: 84.5 on physically-grounded tasks (speaker recognition, seismology, RF fingerprinting) versus 70.0% on semantic tasks (music genre, language recognition). This reveals that principle-driven and scale-driven approaches offer complementary paths: physical principles enable efficient cross-modal transfer while naturally establishing the boundary between physical and semantic understanding.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04074unread
Adaptive Patching Is Harder Than It Looks For Time-Series Forecasting
Federico Zucchi, Yi Xie, Chao Zhang, Keyuan Luo, Thomas Lampert, Ziyue Li · 2026-06-05
arXiv:2606. 04074v1 Announce Type: new Abstract: Adaptive patching is a recent and compelling proposal for time-series Transformers: allocate finer patches where the sequence looks locally informative.
Read next because Adaptive Patching Is Harder Than It Looks For Time-Series Forecasting overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, rect, under, alignment, eval, line, rate, control. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04074v1 Announce Type: new Abstract: Adaptive patching is a recent and compelling proposal for time-series Transformers: allocate finer patches where the sequence looks locally informative. This paper asks under what conditions a content-adaptive patching operator should outperform a tuned uniform one. Local heterogeneity alone is not enough: under pointwise forecasting losses, a complex-looking region is not automatically one where finer patching reduces the loss. We model patching as a budgeted bitrate allocation and derive an explicit threshold that a dynamic patching rule must satisfy to beat a well-tuned uniform baseline, then bound the achievable improvement both locally (a quadratic surrogate) and globally (a strong-convexity bound under the model's assumptions). Two structural results follow: without a coupling constraint, scalar local complexity cannot produce a non-uniform optimum under a common loss landscape; and once the backbone is trained to its representation-aware optimum, the alignment gain collapses around a well-tuned uniform patch size. To test these predictions, we run a controlled isolation study on three representative architectures, replacing each adaptive mechanism with a uniform patch-size sweep while keeping the backbone, data, and training protocol fixed. On standard long-horizon forecasting benchmarks, the validation-selected uniform baseline is competitive with the dynamic counterpart, with per-setting effects concentrated near zero and no consistent directional advantage once results are aggregated by dataset. The larger gains we do observe are method- and dataset-specific. Adaptive patching should therefore be evaluated against a tuned uniform baseline; its value depends on whether a cheap and reliable routing signal can identify where finer patches actually reduce forecasting loss.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04073unread
TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection
Xiancheng Wang, Zhibo Zhang, Ran Li, Rui Wang, Minghang Zhao, Shisheng Zhong, Lin Wang · 2026-06-05
arXiv:2606. 04073v1 Announce Type: new Abstract: This paper proposes a two-stage pseudo anomaly-guided anomaly detection method (\textbf{T}wo-stage \textbf{P}seudo \textbf{A}nomaly-guided \textbf{A}nomaly \textbf{D}etection, \textbf{TPA-AD}) for axle-box bearing time-series anomaly detection (time series anomaly detection, TSAD) under the setting where only normal samples are available for training.
Read next because TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, rate, compare, control, stage, lora, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04073v1 Announce Type: new Abstract: This paper proposes a two-stage pseudo anomaly-guided anomaly detection method (\textbf{T}wo-stage \textbf{P}seudo \textbf{A}nomaly-guided \textbf{A}nomaly \textbf{D}etection, \textbf{TPA-AD}) for axle-box bearing time-series anomaly detection (time series anomaly detection, TSAD) under the setting where only normal samples are available for training. The method first generates pseudo-anomalous windows near the normal boundary using a reconstruction model and per-feature target-error control. It then learns anomaly-sensitive representations through contrastive learning between normal and pseudo-anomalous windows, and finally produces window-level and point-level anomaly scores using k-nearest neighbors (KNN). Compared with existing methods that rely on known fault categories, real anomaly priors, or random anomaly injection, TPA-AD improves the separability of the normal boundary by constructing pseudo-anomalies in boundary neighborhoods and can jointly handle continuous and discrete features in mixed-variable scenarios. The main experiments are conducted on bearing fault detection datasets and degradation-process datasets, with an additional exploratory extension on $13$ public TSAD datasets. The results show that the proposed method yields relatively stable anomaly responses, is sensitive to degradation evolution, and demonstrates a certain degree of broader applicability on public TSAD benchmarks and real high-speed-train-related bearing data.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04058unread
Spectral Scaling Laws of Muon
Gagik Magakyan, Pablo Parrilo, Asuman Ozdaglar · 2026-06-05
arXiv:2606. 04058v1 Announce Type: new Abstract: Orthonormalized update rules have rapidly become a leading choice of optimizer for training large language models, with recent open-source state-of-the-art models adopting Muon.
Read next because Spectral Scaling Laws of Muon overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, source, recipe, without, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04058v1 Announce Type: new Abstract: Orthonormalized update rules have rapidly become a leading choice of optimizer for training large language models, with recent open-source state-of-the-art models adopting Muon. To keep these updates tractable, Muon performs the orthonormalization with the Newton--Schulz (NS) iteration. Since NS is only approximate, directions with small singular values fail to be orthonormalized. In Muon, NS is applied to the momentum matrix at every step, yet little is known about how the singular value spectrum of these momentum matrices behaves during training, or how that behavior changes with model size. We present the first systematic study of this question. Tracking singular value quantiles of the momentum buffer across layers in models ranging from 77M to 2.8B parameters, we observe a consistent picture: after a short burn-in, the quantiles stabilize at a value determined by the layer type and model size. These stabilization values follow remarkably clean power laws in model size, with layer-dependent exponents. Layers up to mid-late depth scale very mildly with model size $M$ (around $M^{-0.25}$), so the standard 5-step NS configuration used at academic scale will continue to orthonormalize them at much larger scales. Some of the late layers, however, scale much more aggressively (up to $M^{-0.96}$) and will fall into the NS failure regime at frontier scale unless one uses more NS iterations or better-tuned coefficients. NS iterations are computationally expensive at scale; our laws give practitioners a principled, layer-aware recipe for choosing the minimum NS configuration that still orthonormalizes the directions that matter -- avoiding unnecessary computation without sacrificing update quality.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses failure.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04053unread
A Goal-Set Characterization of Task Composition in the Boolean Task Algebra
Eduardo Terr\'es-Caballero, Herke van Hoof · 2026-06-05
arXiv:2606. 04053v1 Announce Type: new Abstract: The Boolean Task Algebra (BTA) provides a principled framework for zero-shot task composition in reinforcement learning by equipping goal-reaching tasks with Boolean operations.
Read next because A Goal-Set Characterization of Task Composition in the Boolean Task Algebra overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, control, does, full, position. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04053v1 Announce Type: new Abstract: The Boolean Task Algebra (BTA) provides a principled framework for zero-shot task composition in reinforcement learning by equipping goal-reaching tasks with Boolean operations. We revisit its structural assumptions and formalize a collapse in the space of optimal extended Q-value functions: in deterministic MDPs, every such function is fully determined by the universal and empty tasks. This makes the logarithmic set of base tasks proposed in the original BTA formulation redundant. Building on this observation, we introduce a goal-set-based composition method that performs logical operations on goal sets and reconstructs composed value functions by selecting slices from the universal and empty value functions. This reduces learning costs for standard BTA and reduces composition time for both BTA and Skill Machines, while preserving policy performance. Experiments across tabular, visual, function-approximation, and continuous-control domains show that learning additional base tasks does not yield better performance. Finally, we study the stochastic setting and provide a counterexample showing that this collapse need not hold, that is, optimal composition may require accounting for exponentially many policies in the number of goals. Code is available at https://github.com/EduardoTerres/bta_paper.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses counterexample.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04051unread
RUBAS: Rubric-Based Reinforcement Learning for Agent Safety
Xian Qi Loye, Qinglin Su, Zhexin Zhang, Shiyao Cui, Qi Zhu, Fei Mi, Hongning Wang, Minlie Huang · 2026-06-05
arXiv:2606. 04051v1 Announce Type: new Abstract: The evolution of LLMs into tool-enabled agents creates a new class of safety challenges associated with real-world execution rather than simple text generation.
Read next because RUBAS: Rubric-Based Reinforcement Learning for Agent Safety overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, class, alignment, line, completion, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04051v1 Announce Type: new Abstract: The evolution of LLMs into tool-enabled agents creates a new class of safety challenges associated with real-world execution rather than simple text generation. Existing alignment methods often rely on coarse refusal signals or static supervision, making it difficult to balance safety with useful tool execution across diverse agentic risks. We introduce RUBAS, a rubric-based reinforcement learning framework for agent safety. RUBAS decomposes agent behavior into four dimensions: tool-use safety, argument safety, response safety, and helpfulness. These structured rubrics provide fine-grained and interpretable rewards over complete agent trajectories, enabling reinforcement learning to optimize safe tool use while preserving task completion. Extensive experiments across multiple agent safety benchmarks and models show that RUBAS improves safety over standard alignment baselines, reduces tool-grounded hallucinations, and maintains competitive utility. Our results suggest that multi-dimensional rubric rewards provide an effective training signal for aligning LLM agents in safety-critical tool-use settings.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04033unread
Inverse Critical Experiment Design via Gradient Optimization and a Multigroup Attention-Based Neural Network Architecture
Will Savage, Logan Burnett, Dean Price · 2026-06-05
arXiv:2606. 04033v1 Announce Type: new Abstract: The validation of advanced nuclear reactor designs and fuel concepts requires critical experiments with high neutronic similarity to the target technology.
Read next because Inverse Critical Experiment Design via Gradient Optimization and a Multigroup Attention-Based Neural Network Architecture overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, text, rect, rate, full, trained, position, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04033v1 Announce Type: new Abstract: The validation of advanced nuclear reactor designs and fuel concepts requires critical experiments with high neutronic similarity to the target technology. Neutronic similarity is quantified by the correlation coefficient $c_k$, which captures the shared bias in $k_\text{eff}$ induced by uncertainties in nuclear data. Generally, a $c_k\geq0.9$ is needed for an experiment to be sufficiently similar to a target technology. This work presents a methodology for the inverse design of critical experiments. Deep neural network surrogate modeling and nonparametric gradient optimization are used to generate experiment geometries that maximize $c_k$. A deep neural network is trained on OpenMC-calculated sensitivity vectors for grid-based critical experiment geometries. The model architecture combines a U-Net convolutional encoder-decoder with a novel multigroup attention pooling layer, introduced to capture the differing spatial dependencies of sensitivities. Multigroup attention pooling is shown to achieve better performance than traditional pooling, as well as interpretable internal behavior. The differentiability of the surrogate enables gradient-based optimization of the full combinatorial design space, allowing $c_k$ to be maximized by directly changing the material assignment of each position in the geometry grid. The method is applied to the validation of the TN-Americas TN-LC transportation cask with HALEU fuel, for which existing critical experiment coverage is limited. The optimization procedure is shown to produce experiment geometries achieving $c_k$ scores of 0.97757, 0.81324, and 0.93276 for three configurations of interest. This approach demonstrates the potential of deep learning and gradient optimization to accelerate the development of advanced nuclear technology.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias.
source ↗open
score 100arxiv cs.LG (Machine Learning)arxiv:2606.04028unread
Novel Aspects of IEEE SA P3109 Arithmetic Formats for Machine Learning
Andrew Fitzgibbon, Christoph M. Wintersteiger, Jeffrey Sarnoff · 2026-06-05
arXiv:2606. 04028v1 Announce Type: new Abstract: The IEEE P3109 draft standard defines a parameterized family of binary floating-point formats and associated operations, with a focus on facilitating machine learning.
Read next because Novel Aspects of IEEE SA P3109 Arithmetic Formats for Machine Learning overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: under, width, rate, implement, factor. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04028v1 Announce Type: new Abstract: The IEEE P3109 draft standard defines a parameterized family of binary floating-point formats and associated operations, with a focus on facilitating machine learning. These formats allow efficient and consistent representation of values in a small number of bits. The defined formats are parameterized over width and precision in bits, signedness, and the presence of infinities. Operations are defined by decoding floating-point values to the set of closed extended reals: the reals augmented with positive and negative infinity and NaN (Not a Number). Explicit treatment of NaN and infinite operands ensures that only real arithmetic is invoked in operation definitions. Extensive rounding and saturation modes are defined; stochastic rounding is included. Operations are exception-free, accelerating throughput, with exceptional situations communicated through return values, e.g., NaN. Operations on blocks of values sharing a common scale factor are defined in terms of the underlying operations in a uniform manner. System vendors may describe approximate implementations via a novel scale-invariant measure, akin to units in the last place, called kappa-approximation. Standard function definitions and various other properties are mechanically verified and generated using formal specifications.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses negative.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05797unread
Causal Longitudinal Prior-Fitted Networks for Counterfactual Outcome Prediction
Amirhossein Zare, Amirhessam Zare, Herlock Rahimi, Reza Salarikia, Mohammad Kashkooli · 2026-06-05
arXiv:2606. 05797v1 Announce Type: cross Abstract: Longitudinal treatment decisions require predicting potential outcomes under future treatment sequences in the presence of time-varying confounding, heterogeneous patient dynamics, and limited domain-specific data.
Read next because Causal Longitudinal Prior-Fitted Networks for Counterfactual Outcome Prediction overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, text, under, eval, line, without, trained, test. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05797v1 Announce Type: cross Abstract: Longitudinal treatment decisions require predicting potential outcomes under future treatment sequences in the presence of time-varying confounding, heterogeneous patient dynamics, and limited domain-specific data. Existing longitudinal causal estimators typically train a new model for each cohort or simulator. We introduce Causal Longitudinal Prior-Fitted Networks (CausalLongPFN), a prior-fitted in-context predictor for longitudinal causal prediction. The model is pretrained entirely on synthetic episodes sampled from a broad prior over temporal structural causal models, exposing it to treatment-confounder feedback, latent heterogeneity, nonlinear state evolution, delayed effects, and cumulative treatment responses. At test time, CausalLongPFN is frozen: it conditions on support trajectories, a query history, and a proposed future treatment sequence, and returns a predictive distribution over future outcomes without gradient updates or propensity-model fitting. Multi-step predictions are obtained by recursively applying the one-step predictor under the specified treatment sequence. We evaluate on branchable cancer, HIV, and warfarin benchmarks with ground-truth counterfactual labels, and on factual-only rolling-origin prediction in MIMIC-III ICU trajectories. CausalLongPFN is competitive with domain-trained longitudinal baselines on counterfactual benchmarks and performs strongly on factual MIMIC-III prediction, suggesting that broad synthetic causal pretraining can provide a useful frozen alternative when repeated domain-specific training is costly or impractical.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses confound, benchmark.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05733unread
Zero-Copy Semantic Contagion: An In-Memory Streaming Architecture for Evolving Attention Graphs
Kabir Murjani · 2026-06-05
arXiv:2606. 05733v1 Announce Type: cross Abstract: Per-ticker forecasting models dominate financial time-series work yet remain blind to cross-company propagation: a foundry disruption in Taiwan does not register in a single-asset model until Apple's own price has already moved.
Read next because Zero-Copy Semantic Contagion: An In-Memory Streaming Architecture for Evolving Attention Graphs overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, rect, eval, line, rate, project, does, propagate. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05733v1 Announce Type: cross Abstract: Per-ticker forecasting models dominate financial time-series work yet remain blind to cross-company propagation: a foundry disruption in Taiwan does not register in a single-asset model until Apple's own price has already moved. To address this limitation, we introduce a heterogeneous Rust-Python streaming architecture that maps cross-company attention as a continuous-time graph driven directly from text. We show that on the ingestion side, a zero-copy Rust edge parses news records in $\sim$100 ns and scans the target equity universe in $\sim$1.2 $\mu$s. On the inference end, a multivariate Neural Hawkes Process featuring per-node continuous-time LSTM states and a bilinear latent projection propagates directed excitation, while an adaptive pruning rule bounds the computational cost of dynamic neighborhood updates. Combining these stages, we demonstrate an end-to-end processing latency of $\sim$13 ms per incoming news record on a single commodity CPU. Evaluated on a one-month temporal holdout of the FNSPID corpus (638 articles across 47 tickers), the system delivers a $1.70\times$ precision lift over random at the 90th-percentile next-day return threshold, and $3.36\times$ over a same-sector baseline. Crucially, removing the graph topology collapses precision to zero, confirming that the dynamic attention network is the sole driver of cross-company signal in this architecture.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05599unread
Mitigating the Curse of Dimensionality in Uniform Convergence of Deep Neural Networks via Smooth Activations
Yizhe Ding, Runze Li, Jia Liu, Lingzhou Xue · 2026-06-05
arXiv:2606. 05599v1 Announce Type: cross Abstract: This paper establishes a theoretical framework for the uniform convergence of smoothly activated deep neural network (DNN) estimators.
Read next because Mitigating the Curse of Dimensionality in Uniform Convergence of Deep Neural Networks via Smooth Activations overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, rate, position, contexts, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05599v1 Announce Type: cross Abstract: This paper establishes a theoretical framework for the uniform convergence of smoothly activated deep neural network (DNN) estimators. While standard ReLU networks achieve minimax-optimal rates in the $L^2(P)$ norm for various nonparametric regression tasks, we establish a theoretical lower bound demonstrating that least-squares ReLU estimators can suffer from the curse of dimensionality in their uniform convergence behavior. Motivated by the need for reliable uniform guarantees in downstream tasks requiring worst-case reliability, we address this limitation by analyzing smoothly activated DNNs (smooth DNNs), encompassing both feedforward and residual structures. We establish novel pseudo-dimension bounds, non-asymptotic approximation guarantees, and H\"older-norm bounds for the approximators of these models. Leveraging these results, we derive non-asymptotic uniform convergence rates for smooth DNN estimators across multiple statistical contexts, including Huber, least-squares, quantile, and logistic regression. We prove that smooth DNNs can mitigate the {curse of dimensionality} in uniform convergence by adaptively exploiting the low-dimensional hierarchical composition structure of the target function. Supported by both simulation studies and a real-world application, our results position smooth DNNs as a theoretically grounded and practically viable alternative to ReLU networks for statistical learning tasks requiring uniform guarantees.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05441unread
GOTabPFN: From Feature Ordering to Compact Tokenization for Tabular Foundation Models on High-Dimensional Data
Al Zadid Sultan Bin Habib, Md Younus Ahamed, Prashnna Kumar Gyawali, Gianfranco Doretto, Donald A. Adjeroh · 2026-06-05
arXiv:2606. 05441v1 Announce Type: cross Abstract: We investigate how to make small tabular foundation models effective for High-Dimensional, Low-Sample Size (HDLSS) tabular prediction without retraining large backbones.
Read next because GOTabPFN: From Feature Ordering to Compact Tokenization for Tabular Foundation Models on High-Dimensional Data overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: under, token, line, without, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05441v1 Announce Type: cross Abstract: We investigate how to make small tabular foundation models effective for High-Dimensional, Low-Sample Size (HDLSS) tabular prediction without retraining large backbones. We introduce Graph-guided Ordering with Local Refinement (GO-LR), show its equivalence to weighted Minimum Linear Arrangement, and interpret the practical solver as a TSP-path-style surrogate. We propose GOTabPFN,which builds on GO-LR, and a Neuro-Inspired Subunit Compression (NSC) unit to pool locally adjacent ordered features into meta-features, yielding a compact representation that makes TabPFN-style prediction practical in HDLSS regimes. Across tabular benchmarks, GOTabPFN improves stability and accuracy under tight token budgets.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.06351unread
Function-Space Priors for Bayesian Neural ODEs with Application to Vessel Trajectory Prediction
Jaeyeong Lee, Wonmo Koo, Heeyoung Kim · 2026-06-05
arXiv:2606. 06351v1 Announce Type: new Abstract: Vessel trajectory prediction from Automatic Identification System (AIS) data is essential for maritime situational awareness, yet it remains challenging due to irregular sampling, missing reports, and complex dynamics.
Read next because Function-Space Priors for Bayesian Neural ODEs with Application to Vessel Trajectory Prediction overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, eval, line, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06351v1 Announce Type: new Abstract: Vessel trajectory prediction from Automatic Identification System (AIS) data is essential for maritime situational awareness, yet it remains challenging due to irregular sampling, missing reports, and complex dynamics. Beyond accurate point forecasts, maritime applications also demand well-calibrated uncertainty estimates for reliable decision-making. Bayesian Neural Ordinary Differential Equations (ODEs) offer a principled framework for continuous-time trajectory modeling with uncertainty quantification by placing a prior over the neural vector field parameters. However, the commonly used isotropic Gaussian weight prior fails to encode informative structural properties of vessel dynamics, such as smoothness and locality. Existing function-space Bayesian neural network methods address this limitation for static mappings, but do not transfer directly to Neural ODEs, where the primary quantity of interest is the trajectory rather than the vector field itself. In principle, one could place a Gaussian process (GP) prior directly over ODE solutions, but this requires propagating distributions through a nonlinear ODE solver, which is analytically intractable. To address this challenge, we adopt a practical approach that imposes a GP-kernel-based prior directly on the vector field evaluated at a finite set of measurement points. Specifically, we augment the standard weight-space variational objective with a kernel-based regularizer that penalizes deviations of the vector field from the structure implied by a GP prior. To handle long and irregular AIS trajectories, we further combine this function-space regularization with probabilistic multiple shooting, which decouples inference across temporal segments while maintaining global consistency.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.06342unread
Symmetric Divergence and Normalized Similarity: A Unified Topological Framework for Representation Analysis
Yan Wang, Tianyang Hu · 2026-06-05
arXiv:2606. 06342v1 Announce Type: new Abstract: Topological Data Analysis (TDA) offers a principled, intrinsic lens for comparing neural representations.
Read next because Symmetric Divergence and Normalized Similarity: A Unified Topological Framework for Representation Analysis overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, under, eval, rate, without, symmetry, asymmetry. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.06342v1 Announce Type: new Abstract: Topological Data Analysis (TDA) offers a principled, intrinsic lens for comparing neural representations. However, existing paired topological divergences (e.g., RTD) are limited by heuristic asymmetry and, more critically, unbounded scores that depend on sample size, hindering reliable cross-scenario benchmarking. To address these challenges, we develop a unified topological toolkit serving two complementary needs: fine-grained structural diagnosis and robust, standardized evaluation. First, we complete the RTD framework by introducing Symmetric Representation Topology Divergence (SRTD) and its efficient variant SRTD-lite. Beyond resolving the theoretical asymmetry of prior variants, SRTD consolidates diagnostic information into a single, comprehensive cross-barcode signature. This allows for precise localization of structural discrepancies and serves as an effective optimization objective without the overhead of dual directional computations. Second, to enable reliable benchmarking across heterogeneous settings, we propose Normalized Topological Similarity (NTS). By measuring the rank correlation of hierarchical merge orders, NTS yields a scale-invariant metric bounded between -1 and 1, effectively overcoming the scale and sample-dependence of unnormalized divergences. Experiments across synthetic and real-world deep learning settings demonstrate that our toolkit captures functional shifts in CNNs missed by geometric measures and robustly maps LLM genealogy even under distance saturation, offering a rigorous, topology-aware perspective that complements measures like CKA.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05942unread
EML-CD: Causal Mechanism Recovery via EML Symbolic Trees in Structure Learning
Sota Asanuma · 2026-06-05
arXiv:2606. 05942v1 Announce Type: new Abstract: Neural network (NN)-based nonlinear causal discovery methods recover DAG structure but leave each causal mechanism as a black box.
Read next because EML-CD: Causal Mechanism Recovery via EML Symbolic Trees in Structure Learning overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, under, line, rate, control, full, test, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05942v1 Announce Type: new Abstract: Neural network (NN)-based nonlinear causal discovery methods recover DAG structure but leave each causal mechanism as a black box. Waxman et al. argued that extracting causal mechanisms from NN weights is ill-posed. We propose EML-CD, a framework that integrates the EML operator (capable of composing elementary functions from a single binary operator) into causal structure learning, with interpretable mechanism recovery as the primary objective. EML-CD represents each edge mechanism as a gated EML binary tree and automatically discovers closed-form causal equations. Analytical Jacobians can be directly computed from the output equations, enabling quantitative understanding of causal effects. On real data (Sachs protein signaling, d=11), EML-CD achieves SHD=11.2 +/- 0.4 (5-seed mean; baselines are single deterministic runs), on par with PC/GES within seed variance and below CAM, while attaching closed-form equations to each detected edge (precision 0.756, recall 0.365). In a controlled bivariate test with known mechanisms, EML-CD recovers 10 of 11 elementary function families faithfully (held-out shape correlation >= 0.96; only high-frequency sine is partial). On a symbolic synthetic benchmark, EML-CD attains a substantially lower and more stable held-out mechanism f-MSE than a fixed SINDy dictionary (mean 3.67 vs. 7644, the latter inflated by catastrophic extrapolation on one seed), although its structure recovery (SHD 14.0) only matches the dictionary and stays below specialized optimizers; on the Causal Chambers light-tunnel subset, a depth-2 model improves F1 over linear OLS-BIC (0.444 vs. 0.273).
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05258unread
Harnessing Source Heterogeneity for Cluster-Structured Transfer Learning
Xiaohui Yin, Jun Jin, Shane J. Sacco, Robert H. Aseltine, Kun Chen · 2026-06-05
arXiv:2606. 05258v1 Announce Type: new Abstract: Transfer learning is a natural strategy when a target population has limited data but multiple related auxiliary sources are available.
Read next because Harnessing Source Heterogeneity for Cluster-Structured Transfer Learning overlaps with clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Factor screen for marker implantation + leakage (2^5: system-prompt length, answer-format length, persona-presence, on-policy, marker-only-loss)". Matching terms: source, line, rate, factor, candidate, model, never. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05258v1 Announce Type: new Abstract: Transfer learning is a natural strategy when a target population has limited data but multiple related auxiliary sources are available. A central difficulty is source heterogeneity: auxiliary sources may not be equally useful, and their usefulness may vary in a structured, cluster-like fashion. Existing transfer-learning methods often reduce source selection to a binary informative/non-informative decision, overlooking subgroups of sources with differential transferability. Motivated by a suicide-risk study using data from the Connecticut Hospital Information Management Exchange (CHIME), comprising 636,758 patients across 27 hospitals, we propose Trans-GLMC, a cluster-structured transfer-learning procedure for generalized linear models. The CHIME setting illustrates the core challenge: hospital-specific risk models are unstable because suicide attempts are rare at any single facility, whereas indiscriminate pooling across hospitals can obscure facility-level differences in patient mix and risk profiles. Trans-GLMC first constructs a coefficient-based distance among the target and candidate sources to recover latent source clusters. It then combines global fusion, within-cluster refinement, and target debiasing to produce an estimator that adapts to the detected structure. We establish a non-asymptotic error bound that improves over its unclustered counterpart whenever a meaningful target cluster exists and matches the unclustered rate up to constants otherwise. In simulations and in the CHIME study, Trans-GLMC improves facility-specific prediction, identifies interpretable communities of hospitals with mutual transferability, and recovers clinically coherent suicide-risk factors.
Potential threat/caveat for clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)": this item discusses bias.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05239unread
HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation
Hongfan Gao, Wangmeng Shen, Bin Yang, Jilin Hu · 2026-06-05
arXiv:2606. 05239v1 Announce Type: new Abstract: Diffusion models have demonstrated strong performance in time series modeling due to their ability to progressively capture complex data distributions through iterative denoising.
Read next because HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, source, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05239v1 Announce Type: new Abstract: Diffusion models have demonstrated strong performance in time series modeling due to their ability to progressively capture complex data distributions through iterative denoising. However, existing approaches struggle with frequency-sensitive denoising, high-frequency reconstruction and balancing global trends with local dynamics. To address these limitations, we propose \textbf{HyFAD}, a \textbf{Hy}brid time-frequency \textbf{D}iffusion model with \textbf{F}requency-\textbf{A}ware embedding for time series imputation. Built upon the DDPM paradigm, HyFAD adopts a coupled time-frequency diffusion framework, in which the reverse denoising proceeds sequentially from the time domain to the frequency domain, enabling coarse-to-fine generation. Specifically, the time-domain diffusion process captures low-frequency global trends, while the frequency-domain diffusion process refines high-frequency spectral components. We further introduce a frequency-aware step embedding that exploits the relationship between diffusion steps and spectral components, providing step-dependent spectral guidance and facilitates more accurate band-wise reconstruction. Extensive experiments on multiple benchmark datasets demonstrate that HyFAD achieves state-of-the-art performance. Our source code is available at https://github.com/hongfangao/HyFAD.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, benchmark.
source ↗open
score 100arxiv stat.ML (Machine Learning)arxiv:2606.05230unread
Central Description Length (CDL) Clustering Validation Index
Mahdi Shamsi, Soosan Beheshti · 2026-06-05
arXiv:2606. 05230v1 Announce Type: new Abstract: Selecting a clustering algorithm and its hyperparameters without labels is a common difficulty in engineering machine learning pipelines that work with unsupervised analysis of sensor, image, or process data.
Read next because Central Description Length (CDL) Clustering Validation Index overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, eval, line, implement, without, does, length, stage. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05230v1 Announce Type: new Abstract: Selecting a clustering algorithm and its hyperparameters without labels is a common difficulty in engineering machine learning pipelines that work with unsupervised analysis of sensor, image, or process data. Clustering validation indices (CVIs) provide internal scores for ranking candidate clusterings, but most popular CVIs are built from Euclidean compactness and separation terms and so tend to favour compact, convex partitions. Their performance is known to degrade on non convex, irregular, or variable density data, where kernel transformations or alternative distance measures are typically used at the cost of additional tuning and computation. This paper introduces the Central Description Length (CDL) clustering validation index. CDL uses the observed within cluster compactness, the estimated cluster centers, and the estimated cluster covariances to compute a probabilistic upper bound on the description length associated with the unobservable true cluster centers. The bound condenses intra cluster compactness and centroid displacement into a single computable quantity and is evaluated on the partition produced by any clustering algorithm. The implementation uses only observable quantities (the data, the partition, the estimated centers, and the estimated covariances) and does not use ground truth labels. On synthetic benchmarks with non convex and arbitrary shape clusters, CDL-CVI selected the reference number of clusters more often and reached higher Adjusted Rand Index (ARI) values than the conventional CVIs we tested, without an additional kernel preprocessing stage. On image benchmarks (MNIST, CIFAR-10, STL-10) clustered from frozen unsupervised embeddings, CDL-CVI returned cluster numbers close to the reference class counts across K-means, DBSCAN, and spectral clustering in the reported trials.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05986unread
AttackPathGNN: Cross-function vulnerability detection in smart contracts using state interference graphs and conjunction pooling
Gabriela Dobrita, Simona-Vasilica Oprea, Adela Bara · 2026-06-05
arXiv:2606. 05986v1 Announce Type: new Abstract: Existing learning-based detectors for Solidity smart-contracts reduce vulnerability detection to syntactic pattern matching within single functions, yet many of the most consequential exploits (The DAO, Cream Finance) exist not in any individual function but in the relationship between functions and in the combination of conditions that made the attack feasible.
Read next because AttackPathGNN: Cross-function vulnerability detection in smart contracts using state interference graphs and conjunction pooling overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, rate, control, emit, test, never. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05986v1 Announce Type: new Abstract: Existing learning-based detectors for Solidity smart-contracts reduce vulnerability detection to syntactic pattern matching within single functions, yet many of the most consequential exploits (The DAO, Cream Finance) exist not in any individual function but in the relationship between functions and in the combination of conditions that made the attack feasible. Thus, we propose AttackPathGNN, a graph neural network (GNN) that reframes detection as reasoning over explicit attack paths. Two architectural choices distinguish it from prior GNN-based detectors: (1)a State Interference Graph that links every pair of functions sharing mutable storage through typed, weighted edges and through directed reentrancy-path edges defined by an explicit five-condition predicate; (2)conjunction pooling, a differentiable AND-aggregator over eight named exploit preconditions whose log-sigmoid form causes the per-function exploit score to collapse whenever any single mitigation (a reentrancy guard, an access-control modifier or SafeMath) is in place. Across five independent training runs, AttackPathGNN attains 92.3+/-0.2% F1 on the SmartBugs Wild held-out test partition (4.3+/-0.3% false-negative rate, 90.8+/-2.5% detection rate on the independently human-labelled SmartBugs Curated benchmark), recovering 6/10 DASP10 categories at 100% on every seed and Reentrancy at 98.7+/-1.8%. Each prediction is emitted with a structured remediation report, turning each verdict into an actionable, function-level audit finding.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses negative, benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05902unread
PriSrv+: Privacy and Usability-Enhanced Wireless Service Discovery with Fast and Expressive Matchmaking Encryption
Yang Yang, Guomin Yang, Yingjiu Li, Pengfei Wu, Rui Shi, Minming Huang, Jian Weng, HweeHwa Pang, Robert H. Deng · 2026-06-05
arXiv:2606. 05902v1 Announce Type: new Abstract: Service discovery is a fundamental process in wireless networks, enabling devices to find and communicate with services dynamically, and is critical for the seamless operation of modern systems like 5G and IoT.
Read next because PriSrv+: Privacy and Usability-Enhanced Wireless Service Discovery with Fast and Expressive Matchmaking Encryption overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, source, rate, compare, control, trained. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05902v1 Announce Type: new Abstract: Service discovery is a fundamental process in wireless networks, enabling devices to find and communicate with services dynamically, and is critical for the seamless operation of modern systems like 5G and IoT. This paper introduces PriSrv+, an advanced privacy and usability-enhanced service discovery protocol for modern wireless networks and resource-constrained environments. PriSrv+ builds upon PriSrv (NDSS'24), by addressing critical limitations in expressiveness, privacy, scalability, and efficiency, while maintaining compatibility with widely-used wireless protocols such as mDNS, BLE, and Wi-Fi. A key innovation in PriSrv+ is the development of Fast and Expressive Matchmaking Encryption (FEME), the first matchmaking encryption scheme capable of supporting expressive access control policies with an unbounded attribute universe, allowing any arbitrary string to be used as an attribute. FEME significantly enhances the flexibility of service discovery while ensuring robust message and attribute privacy. Compared to PriSrv, PriSrv+ optimizes cryptographic operations, achieving 7.62* faster for encryption and 6.23* faster for decryption, and dramatically reduces ciphertext sizes by 87.33%. In addition, PriSrv+ reduces communication costs by 87.33% for service broadcast and 86.64% for anonymous mutual authentication compared with PriSrv. Formal security proofs confirm the security of FEME and PriSrv+. Extensive evaluations on multiple platforms demonstrate that PriSrv+ achieves superior performance, scalability, and efficiency compared to existing state-of-the-art protocols.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05844unread
GenTI: Benchmarking LLMs for Autonomous IDPS Rule Generation for Unseen Attacks
Hassan Jalil Hadi, Rehana Yasmin, Ali Shoker · 2026-06-05
arXiv:2606. 05844v1 Announce Type: new Abstract: Rule-based Intrusion Detection and Prevention Systems (IDPS) offer precise attack detection as well as mitigation, however their manually crafted, signature-driven rules limit adaptability to emerging and zero-day threats.
Read next because GenTI: Benchmarking LLMs for Autonomous IDPS Rule Generation for Unseen Attacks overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, class, eval, line, rate, chain. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05844v1 Announce Type: new Abstract: Rule-based Intrusion Detection and Prevention Systems (IDPS) offer precise attack detection as well as mitigation, however their manually crafted, signature-driven rules limit adaptability to emerging and zero-day threats. Additionally, existing public datasets (e.g., CICIDS2017, UNSW-NB15) focus on traffic classification and provide little structured information to support automatic rule synthesis or prevention logic. To address this gap, we propose Generative Thread Intelligence (GenTI) \footnote{GenTI refers to the proposed framework, and GTI refers to the dataset.} an LLM-driven benchmark for automatic generation of IDPS rules targeting unseen attacks. The dataset (GTI) aggregates over 150k detection and prevention rules from Snort, Suricata, Emerging Threats, as well as 50k YARA, each annotated with protocol behavior, payload signatures, contextual relationships, mappings to Cyber Threat Intelligence (CTI), along with actionable response types (alert, drop, reject). Moreover, on top of this corpus we design an LLM-based pipeline that transforms analyst prompts and representative payloads into deployable rules via structured prompt engineering, Chain-of-Thought (CoT) reasoning, as well as a Chain-of-Verification (CoVe) loop for syntactic, semantic, and security validation. The generated rules are executed in real time on (Snort/Suricata) and evaluated by syntax accuracy, semantic similarity, CTI coverage, security effectiveness as well as unseen attacks detection. Furthermore, our GenTI instantiation achieves a composite rule-quality score of 89.4\%, with 94.8\% CTI coverage, improving unseen attacks detection from 45\% to 87.4\% and reducing the false-positive rate from 8.5\% to 2.3\%. Overall, GenTI establishes the first large-scale benchmark that tightly couples rule-level CTI with LLM-based automation, enabling adaptive, self-evolving IDPS.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05821unread
PriSrv: Privacy-Enhanced and Highly Usable Service Discovery in Wireless Communications
Yang Yang, Robert H. Deng, Guomin Yang, Yingjiu Li, HweeHwa Pang, Minming Huang, Rui Shi, Jian Weng · 2026-06-05
arXiv:2606. 05821v1 Announce Type: new Abstract: Service discovery is essential in wireless communications.
Read next because PriSrv: Privacy-Enhanced and Highly Usable Service Discovery in Wireless Communications overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone". Matching terms: eval, rate, control, leaking. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05821v1 Announce Type: new Abstract: Service discovery is essential in wireless communications. However, existing protocols provide limited privacy protection, leaking sensitive device information and opening routes to network attacks. This paper proposes a private service discovery protocol, called PriSrv, which enables both service providers and clients to specify fine-grained authentication policies before establishing connections. PriSrv achieves this via a dual-layer matching architecture: an outer layer filters mismatched entities using public attributes, while an inner layer handles mutual authentication using selectively disclosed private attributes. As a core component, we introduce the primitive of anonymous credential-based matchmaking encryption (ACME), which enables dual-layer matching in a single step to achieve bilateral policy control, selective attribute disclosure, and multi-show unlinkability. To instantiate ACME, we design a fast anonymous credential (FAC) scheme providing constant-size credentials and efficient verification. We demonstrate PriSrv's interoperability by integrating it with popular wireless frameworks including EAP, mDNS, BLE, and AirDrop. Detailed formal security proofs and extensive performance evaluations across desktop, laptop, smartphone, and Raspberry Pi platforms demonstrate that PriSrv provides enhanced privacy guarantees with high usability, achieving secure discovery in less than one second on mainstream mobile devices.
Potential threat/caveat for clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05787unread
SentinelRAG: Synthetic Sentinel Knowledge for RAG Database Copyright Protection
Tsun On Kwok, Xi Yang, Ki Sen Hung, Chang Liu, Yangqiu Song · 2026-06-05
arXiv:2606. 05787v1 Announce Type: new Abstract: Protecting proprietary RAG databases from unauthorized redistribution is challenging: existing watermarking methods either inject fabricated relations between real entities, polluting the knowledge base with misinformation, or embed fragile lexical patterns that adversarial paraphrasing easily removes.
Read next because SentinelRAG: Synthetic Sentinel Knowledge for RAG Database Copyright Protection overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Implement Chen et al. persona-vector extraction recipe and compare to project's centroid-difference recipe", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation". Matching terms: rate, compare, test. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05787v1 Announce Type: new Abstract: Protecting proprietary RAG databases from unauthorized redistribution is challenging: existing watermarking methods either inject fabricated relations between real entities, polluting the knowledge base with misinformation, or embed fragile lexical patterns that adversarial paraphrasing easily removes. We propose SentinelRAG, a watermarking framework that embeds style-consistent but fictitious knowledge entries into the RAG database. Our key insight is that synthetic knowledge describing fictitious entities is unlikely to be retrieved by legitimate queries, yet can be reliably triggered through targeted probes known only to the data owner. Experiments on four datasets ranging from 2.9k to 8.8M documents demonstrate that SentinelRAG achieves statistically significant detection $p < 10^{-5}$ across all tested configurations at only a 0.1% injection rate. Compared to the state-of-the-art, our method significantly reduces the false detection rate while maintaining negligible interference with legitimate user queries.
Potential threat/caveat for clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)": this item discusses adversarial.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05779unread
TinyML-Driven Cybersecurity for Autonomous Spacecraft: Latency-Accuracy Analysis for SPARTA RF and Cyber Threat Detection
Van Le, Trevor Tran, Tan Le · 2026-06-05
arXiv:2606. 05779v1 Announce Type: cross Abstract: Autonomous spacecraft require rapid, lightweight, and reliable onboard detection of cyber-RF threats.
Read next because TinyML-Driven Cybersecurity for Autonomous Spacecraft: Latency-Accuracy Analysis for SPARTA RF and Cyber Threat Detection overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, class, line, rate, model. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05779v1 Announce Type: cross Abstract: Autonomous spacecraft require rapid, lightweight, and reliable onboard detection of cyber-RF threats. Using the SPARTA attack model, we analyze the latency-accuracy trade-offs of TinyML-compatible classical models -- Random Forest, Logistic Regression, SVM, and MLP -- for detecting uplink jamming, Fake-NR spoofing, payload manipulation, ground-segment compromise, and unauthorized command injection. We present a physics-informed theoretical analysis of each model's computational complexity, VC dimension, Lipschitz continuity, and latency scaling, supported by empirical measurements on adversarial RF spectrograms generated via BandErasure, FakeNR, and NoiseBurst corruption modes. Results show that Logistic Regression achieves microsecond-level inference with only a 1\% accuracy drop relative to Random Forest, making it an effective TinyML baseline for onboard autonomy. The study also identifies opportunities for advancing spacecraft cybersecurity through richer feature encoders and multi-timescale learning architectures, building on recent progress in edge intelligence and trustworthy AI.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses adversarial.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05725unread
An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Traffic
Shuze Liu, Qianwen Guo, Yushun Dong · 2026-06-05
arXiv:2606. 05725v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed through hosted APIs, making model extraction a practical threat to model ownership and service security.
Read next because An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Traffic overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, eval, line, rate, extraction, compare, test. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05725v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed through hosted APIs, making model extraction a practical threat to model ownership and service security. However, individual extraction queries often resemble benign requests, and existing evaluations often focus on single-query anomaly scoring or pure benign-versus-attacker user settings. We formulate model extraction monitoring as benign-calibrated traffic-window distribution testing and show that an embarrassingly simple detector is effective: embed incoming queries into a semantic space and test whether their aggregate distribution deviates from historical benign traffic. We instantiate the detector with maximum mean discrepancy (MMD), using only benign-vs-benign comparisons to set the decision threshold. We evaluate on fourteen attacker-normal query pairs from four extraction scenarios and compare with adapted PRADA, SEAT, CAP, DATE, and marginal Mahalanobis baselines. Across three random seeds, MMD achieves 0.3% benign FPR, 100.0% pure-attacker TPR, 90.5% average TPR over attacker fractions, and 95.1% balanced accuracy. These results show that benign-calibrated distribution testing is a strong empirical baseline for model extraction detection in both user-level and mixed multi-user LLM API traffic. Code is released at: https://github.com/LabRAI/mmd-llm-mea-detection.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05714unread
Hybrid CNN-LSTM Framework for Intelligent Cyber Attack Detection and Prevention in U.S. Critical Digital Infrastructure: A Comparative Machine Learning Evaluation on CSE-CIC-IDS2018
Md. Iqbal Hossan, Md. Serajul Kabir Chowdhury Rubel, Md. Arifur Rahman, B. M. Taslimul Haque · 2026-06-05
arXiv:2606. 05714v1 Announce Type: new Abstract: Digital infrastructure is growing at a rapid pace in the United States, and as a result, exposure to advanced cyber threats to critical sectors including healthcare, finance, transportation, energy and government systems is growing.
Read next because Hybrid CNN-LSTM Framework for Intelligent Cyber Attack Detection and Prevention in U.S. Critical Digital Infrastructure: A Comparative Machine Learning Evaluation on CSE-CIC-IDS2018 overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, eval, implement, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05714v1 Announce Type: new Abstract: Digital infrastructure is growing at a rapid pace in the United States, and as a result, exposure to advanced cyber threats to critical sectors including healthcare, finance, transportation, energy and government systems is growing. The traditional cybersecurity approaches, including signature-based intrusion detection systems, have become less effective against today's cyber attacks, as they are unable to detect unknown and changing attacks in real time. To overcome these constraints, this research suggests a smart cyber-defense system, which utilizes Artificial Intelligence (AI) and Machine Learning (ML) algorithms in the detection and prevention of cyber attacks in the U.S. digital infrastructure. This study uses the CSE-CIC-IDS2018 dataset, which is a realistic network traffic dataset, along with various cyber attack scenarios, including Distributed Denial of Service (DDoS), brute force attacks, botnets, infiltration attacks, and web-based attacks. A number of machine learning and deep learning models such as Random Forest, XGBoost, Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks are implemented and evaluated to be used in identifying malicious network behavior and boosting the accuracy of intrusion detection. The framework proposed combines data preprocessing, feature engineering, real-time traffic monitoring, intelligent threat classification with automated prevention mechanisms to build cybersecurity resilience. E
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05701unread
Cognitive Threat Intelligence and Explainable Federated Security Analytics for distributed Infrastructure Systems
Md. Arifur Rahman, B. M. Taslimul Haque, Md. Iqbal Hossan, Md. Serajul Kabir Chowdhury Rubel · 2026-06-05
arXiv:2606. 05701v1 Announce Type: new Abstract: The increasing adoption of distributed infrastructure systems, cloud computing, Internet of Things (IoT) technologies, and edge-based architectures has significantly expanded the cybersecurity attack surface and introduced increasingly sophisticated cyber threats.
Read next because Cognitive Threat Intelligence and Explainable Federated Security Analytics for distributed Infrastructure Systems overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rate, trained, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05701v1 Announce Type: new Abstract: The increasing adoption of distributed infrastructure systems, cloud computing, Internet of Things (IoT) technologies, and edge-based architectures has significantly expanded the cybersecurity attack surface and introduced increasingly sophisticated cyber threats. Conventional centralized intrusion detection approaches often face challenges related to scalability, data privacy, communication overhead, and limited transparency in artificial intelligence-driven decision-making processes. To address these limitations, this study proposes a Cognitive Threat Intelligence and Explainable Federated Security Analytics framework for distributed infrastructure systems. The proposed framework integrates Federated Learning (FL), Explainable Artificial Intelligence (XAI), and cognitive cybersecurity analytics to enable collaborative and privacy-preserving cyber threat detection across distributed network environments. Instead of transmitting sensitive raw network traffic data to centralized servers, local security models are independently trained at distributed nodes, where only encrypted model parameters and updates are shared through a federated aggregation mechanism. This decentralized learning architecture improves privacy protection while reducing communication dependency and centralized security risks. To enhance intelligent threat analysis, the framework incorporates machine learning and deep learning algorithms including Random Forest, XGBoost, Autoencoder
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05609unread
SlotGCG: Exploiting the Positional Vulnerability in LLMs for Jailbreak Attacks
Seungwon Jeong, Jiwoo Jeong, Hyeonjin Kim, Yunseok Lee, Woojin Lee · 2026-06-05
arXiv:2606. 05609v1 Announce Type: new Abstract: As large language models (LLMs) are widely deployed, identifying their vulnerability through jailbreak attacks becomes increasingly critical.
Read next because SlotGCG: Exploiting the Positional Vulnerability in LLMs for Jailbreak Attacks overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, token, line, rate, implement, position, candidate. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05609v1 Announce Type: new Abstract: As large language models (LLMs) are widely deployed, identifying their vulnerability through jailbreak attacks becomes increasingly critical. Optimization-based attacks like Greedy Coordinate Gradient (GCG) have focused on inserting adversarial tokens to the end of prompts. However, GCG restricts adversarial tokens to a fixed insertion point (typically the prompt suffix), leaving the effect of inserting tokens at other positions unexplored. In this paper, we empirically investigate \emph{slots}, i.e., candidate positions within a prompt where tokens can be inserted. We find that vulnerability to jailbreaking is highly related to the selection of the \emph{slots}. Based on these findings, we introduce the \textit{Vulnerable Slot Score} (VSS) to quantify the positional vulnerability to jailbreaking. We then propose SlotGCG, which evaluates all slots with VSS, selects the most vulnerable slots for insertion, and runs a targeted optimization attack at those slots. Our approach provides a position-search mechanism that is attack-agnostic and can be plugged into any optimization-based attack, adding only 200ms of preprocessing time. Experiments across multiple models demonstrate that SlotGCG significantly outperforms existing methods. Specifically, it achieves 14\% higher Attack Success Rates (ASR) over GCG-based attacks, converges faster, and shows superior robustness against defense methods with 42\% higher ASR than baseline approaches. Our implementation is available at \href{https://github.com/youai058/SlotGCG}{https://github.com/youai058/SlotGCG}
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness, adversarial.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05594unread
The Coverage Gap: Chile's Cyber Disclosure Framework versus the USA, EU and UK
David Mellafe Z · 2026-06-05
arXiv:2606. 05594v1 Announce Type: new Abstract: We introduce the Coverage Gap as a measurable distance between the observable public exposure of critical-infrastructure operators and their declared capability to coordinate vulnerability disclosure.
Read next because The Coverage Gap: Chile's Cyber Disclosure Framework versus the USA, EU and UK overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, directive, under, source, compare, binding, full, stage. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05594v1 Announce Type: new Abstract: We introduce the Coverage Gap as a measurable distance between the observable public exposure of critical-infrastructure operators and their declared capability to coordinate vulnerability disclosure. We instantiate it against the 915 Chilean Operadores de Importancia Vital (OIVs -- Operators of Vital Importance) designated by the National Cybersecurity Agency (ANCI) under Ley 21.663 (Resolucion Exenta No. 87, 16 December 2025). Using a passive-only, OSINT-based method consistent with the principles of ISO/IEC 29147:2018 and Chile's computer-crimes safe harbour (Ley 21.459), we conduct a full-universe census of the foundational disclosure-capability layer (Layer 1, verifiable disclosure contact) across approximately 98.7% of the official catalogue. Only 16 of 915 OIVs (1.7%) publish a verifiable RFC 9116 disclosure channel; among operators of physical-world infrastructure -- energy, health, banking, telecommunications, fuel, water, transport, and state administration -- fewer than ten do so, and all four major banks and both telecommunications incumbents lack one entirely. This compares with over 99% adherence in the U.S. federal civilian branch under CISA Binding Operational Directive 18-01. Email-authentication misconfiguration affects 766 of 915 (84%) OIVs, and end-of-life or known-vulnerable stack components an estimated 23.5% (Wilson 95% CI [12%, 38%]). Cross-jurisdictional benchmarking situates Chile roughly eight years behind the USA, the UK, and the Netherlands on email-authentication mandates, and three years behind Denmark. We propose a four-stage roadmap modelled on BOD 18-01 and the UK Public-Sector DMARC Toolkit, and release the open-source tool anci-oiv-resolver (Apache 2.0) to enable independent reproduction of the OIV-domain mapping that underpins universe-scale auditing.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05584unread
Dimensionality Reduction for Cyberattack Classification: A Comparative Evaluation of PCA and Linear Predictive Coding
Nelly Elsayed, Zag ElSayed, Navid Asadizanjani · 2026-06-05
arXiv:2606. 05584v1 Announce Type: new Abstract: High-dimensional feature representations are widely used in machine learning-based cyberattack detection systems.
Read next because Dimensionality Reduction for Cyberattack Classification: A Comparative Evaluation of PCA and Linear Predictive Coding overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, under, eval, source, line, rate, trained, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05584v1 Announce Type: new Abstract: High-dimensional feature representations are widely used in machine learning-based cyberattack detection systems. However, they increase computational complexity and may hinder deployment in resource-constrained environments. In this paper, we investigate feature compression techniques for cyberattack classification by comparing two dimensionality reduction approaches: Principal Component Analysis (PCA) and Linear Predictive Coding (LPC). Compressed feature representations with varying dimensionalities are generated and evaluated across several classification models. Experimental analysis demonstrates that PCA preserves classification performance even under aggressive compression. On the other hand, LPC provides competitive predictive representations with slightly larger performance degradation. The results show that substantial reductions in feature dimensionality can be achieved with minimal impact on classification accuracy, highlighting the potential of lightweight feature compression for efficient cybersecurity analytics.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05567unread
ZERO-APT: A Closed-Loop Adversarial Framework for LLM-Driven Automated Penetration Testing under Intelligent Defense
Anlan Zheng, Tiantian Zhu · 2026-06-05
arXiv:2606. 05567v1 Announce Type: new Abstract: LLM-driven automated penetration testing agents are typically evaluated against static targets that neither detect nor respond to attacks, so their behavior under intelligent defense remains untested.
Read next because ZERO-APT: A Closed-Loop Adversarial Framework for LLM-Driven Automated Penetration Testing under Intelligent Defense overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, under, eval, rate, chain, emit, test. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05567v1 Announce Type: new Abstract: LLM-driven automated penetration testing agents are typically evaluated against static targets that neither detect nor respond to attacks, so their behavior under intelligent defense remains untested. The causal consistency of multi-step attack chains likewise hinges on unstable LLM reasoning, and agent decisions remain opaque to human analysts. These three shortcomings, in realism, consistency, and auditability, are usually patched in isolation. We present ZERO-APT, a turn-based attacker-defender-judge framework that addresses them within a single architecture. For realism, ZERO-APT embeds a configurable LLM Defender that consumes Sysmon telemetry and detects attacks in real time, exposing the attacker to a live opponent rather than a passive target. For consistency, three architectural mechanisms move causal consistency from unstable LLM reasoning into enforced system architecture: separation of planning from execution, multi-dimensional ReAct feedback, and a hard-constraint-filtered action library. For auditability, a dedicated Judge agent adjudicates each round, maintains global state, and emits structured post-hoc CTI reports that make every decision traceable. We evaluate a Windows Server 2022 post-exploitation prototype across five scenarios with three Defender configurations. ZERO-APT reaches 79\% attack success rate (Aurora 22\%, PentestGPT 39\%), a Causal Consistency Score of 0.860 (Aurora 0.930, Claude Code 0.520), and end-to-end decision auditability through structured CTI reports. We release the benchmark to support evaluation of penetration agents under intelligent defense.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses adversarial, evaluation, benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05503unread
Bitcoin After Block Rewards
Junhyuk Lee · 2026-06-05
arXiv:2606. 05503v1 Announce Type: new Abstract: Bitcoin's block reward is scheduled to decline to zero, raising concerns about whether the network can remain secure once miners rely solely on transaction fees.
Read next because Bitcoin After Block Rewards overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: under, eval, line, compare, does, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05503v1 Announce Type: new Abstract: Bitcoin's block reward is scheduled to decline to zero, raising concerns about whether the network can remain secure once miners rely solely on transaction fees. This paper seeks to identify the conditions under which large-scale and persistent deviation from honest mining can arise. We analyze and compare the payoffs of honest and deviating miners in a sequential decision model, and identify a deviation threshold $G_t$ at which honest mining ceases to be privately optimal. Around the 2024 Bitcoin halving, we show that current mining behavior does not exhibit large-scale or structural deviation. However, when the block reward is removed, the $G_t$ criterion implies that deviation can arise even with a very small fraction of transaction fees. Finally, we evaluate three protocol-level mechanisms: Base Fee, Fee Floor, and an adaptive maximum block size rule, and show that their combination raises the deviation threshold and mitigates incentive breakdown in a fee-only regime. These results provide a practical benchmark for assessing Bitcoin's security as block rewards disappear.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05423unread
Policy-Compliant Cloud Storage Systems
Dimitrios Stavrakakis, Masanori Misono, Julian Pritzi, Harshavardhan Unnibhavi, Nuno Santos, Pramod Bhatotia · 2026-06-05
arXiv:2606. 05423v1 Announce Type: new Abstract: Privacy regulations such as the General Data Protection Regulation (GDPR) impose strict requirements on how personal data is stored, processed, and audited.
Read next because Policy-Compliant Cloud Storage Systems overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, persona, eval, middle, implement, without, language, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05423v1 Announce Type: new Abstract: Privacy regulations such as the General Data Protection Regulation (GDPR) impose strict requirements on how personal data is stored, processed, and audited. While key-value stores (KVS) are widely used in latency-sensitive applications, their simple data model and untrusted cloud deployment environments make GDPR compliance particularly challenging. Existing approaches require invasive code modifications, impose high performance overheads, or overlook the integrity of compliance mechanisms themselves. This paper presents GDPRuler, a trusted middleware system that enables verifiable GDPR compliance for KVS on untrusted clouds without modifying their codebase. GDPRuler deploys a trusted GDPR monitor inside a Confidential Virtual Machine (CVM), which enforces GDPR policies, manages compliance metadata, and maintains tamper-evident audit logs. A declarative policy language translates core GDPR obligations into enforceable runtime rules. To ensure efficiency, GDPRuler encodes metadata compactly within KV records, builds dedicated metadata indexes for GDPR-specific queries, and logs only compliance-relevant events in a space-efficient format. We implement GDPRuler as a transparent proxy for unmodified Redis and RocksDB deployments. Evaluation with YCSB and GDPR-inspired workloads shows that GDPRuler enforces core compliance guarantees with low overheads: GDPRuler achieves ~61% of native KVS throughput with the CVM environment contributing 28%-32% of it, metadata storage overhead remains below 20%, and GDPR queries benefit from 13-182x speedup through metadata indexing. By embedding verifiable policy enforcement into a trusted middleware layer, GDPRuler offers a practical path toward GDPR-compliant KVS on untrusted cloud infrastructures.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05241unread
Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation
Yongjie Wang, Xinyue Zhang, Kunhong Yao, Zhiwei Zeng, Kaisong Song, Jun Lin, Zhiqi Shen · 2026-06-05
arXiv:2606. 05241v1 Announce Type: new Abstract: Public benchmarks enable fair and reproducible evaluation of LLM reasoning, but they become fragile for deep research agents that actively search the web during inference.
Read next because Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, control, leakage. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05241v1 Announce Type: new Abstract: Public benchmarks enable fair and reproducible evaluation of LLM reasoning, but they become fragile for deep research agents that actively search the web during inference. Such agents may retrieve public benchmark metadata, question context, or even ground-truth answers via web search. This gives rise to Search-Time Contamination (STC), where external retrieval bypasses intended reasoning and inflates measured performance. We systematically study STC in deep research agent evaluation. We define three contamination types with increasing severity, namely Benchmark Metadata Leakage, Question-Context Leakage, and Explicit Answer Leakage, and develop detection algorithms to identify them and quantify their impact on agent performance. Evaluating modern deep research agents on six public benchmarks, we find that STC is widespread and can inflate performance by up to 4%. Our findings show that existing evaluations may overestimate true reasoning ability. We therefore advocate contamination-aware practices, including isolated sandboxes, transparent search trajectories, and controlled benchmark access.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.CR (Cryptography and Security)arxiv:2606.05233unread
Domain-Conditioned Safety in Frontier Computer-Using Agents: A 793-Episode Browser Benchmark, a Coding-Domain Cross-Reference, and a Reproducibility Audit of Recent Red-Teaming
Nicholas Saban · 2026-06-05
arXiv:2606. 05233v1 Announce Type: new Abstract: Recent computer-using-agent (CUA) red-teaming papers report prompt-injection attack success rates (ASR) of 42-98%, but these headline numbers cluster on retired models and on the most-vulnerable model in each paper's panel.
Read next because Domain-Conditioned Safety in Frontier Computer-Using Agents: A 793-Episode Browser Benchmark, a Coding-Domain Cross-Reference, and a Reproducibility Audit of Recent Red-Teaming overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, latin, line, rate, without, does, system-prompt, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05233v1 Announce Type: new Abstract: Recent computer-using-agent (CUA) red-teaming papers report prompt-injection attack success rates (ASR) of 42-98%, but these headline numbers cluster on retired models and on the most-vulnerable model in each paper's panel. We ask whether those techniques, reproduced as hand-crafted templates, still work against current frontier CUAs. We release CUA-HandCrafted, a public benchmark of 793 episodes spanning 24 multi-step web tasks, 56 attack templates, 8 attack families, and 4 system-prompt configurations. Against Claude Sonnet 4.6 and GPT-5.4 we measure 0/140 multi-step attack success (Clopper-Pearson 95% upper bound 2.60%); a prompt ablation shows this resistance lives in the model weights. Yet it does not generalize: on a sister coding-agent benchmark (SkillBench), the same weights fall to hand-crafted skill-injection at up to 100%. We argue that the literature's high ASR is largely attributable to RL-optimized injection text rather than the attack categories, and that frontier safety hardening is domain-conditioned, specific to the heavily-targeted browser surface. Reporting techniques without releasing the optimized strings, or extrapolating browser-domain safety to other CUA modalities, makes published ASR numbers unreproducible.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05563unread
SoCRATES: Towards Reliable Automated Evaluation of Proactive LLM Mediation across Domains and Socio-cognitive Variations
Taewon Yun, Hyeonseong Park, Jeonghwan Choi, Hayoon Park, Yeeun Choi, Hwanjun Song · 2026-06-06
arXiv:2606. 05563v1 Announce Type: new Abstract: Evaluating LLM mediators remains challenging, as mediation unfolds as a real-time trajectory shaped by disputants' shifting emotions, intentions, and context.
Read next because SoCRATES: Towards Reliable Automated Evaluation of Proactive LLM Mediation across Domains and Socio-cognitive Variations overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, text, under, alignment, eval, line, rate, length. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05563v1 Announce Type: new Abstract: Evaluating LLM mediators remains challenging, as mediation unfolds as a real-time trajectory shaped by disputants' shifting emotions, intentions, and context. Existing testbeds rely on a few expert-authored domains, vary mainly strategic posture, and score every turn against every topic, introducing off-topic noise. We introduce SoCRATES, a benchmark for evaluating proactive LLM mediators in realistic, multi-domain testbeds. It constructs scenarios from real conflicts through an agentic pipeline across eight domains, probes five socio-cognitive adaptation axes (strategic posture, party composition, history length, emotional reactivity, and cultural identity), and scores each topic only on the turns that advance it via a topic-localized evaluator. The evaluator reaches 0.82 alignment with human experts, more than doubling a per-turn baseline. Benchmarking eight frontier LLMs, we find that even the strongest mediator closes only about a third of the unmediated consensus gap under diverse and realistic testbeds, with performance varying sharply by socio-cognitive axis, highlighting that progress lies in social adaptation to diverse conditions.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05532unread
Individual Gain, Collective Loss: Metacognitive Adaptation in AI-Assisted Creativity
Anna Mikeda · 2026-06-06
arXiv:2606. 05532v1 Announce Type: new Abstract: Recent studies reveal a paradox: AI enhances individual creative outputs while reducing collective diversity.
Read next because Individual Gain, Collective Loss: Metacognitive Adaptation in AI-Assisted Creativity overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: under, eval, rate, control, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05532v1 Announce Type: new Abstract: Recent studies reveal a paradox: AI enhances individual creative outputs while reducing collective diversity. Current explanations -- cognitive offloading and over-reliance -- identify symptoms but not mechanisms. We propose selective metacognitive adaptation: routine AI use redistributes rather than uniformly diminishes metacognitive effort. Some capacities are amplified (partner modeling, surface control), while others are systematically under-supported (originality evaluation, reflective integration). This redistribution explains both individual satisfaction and collective convergence. We present a taxonomy of six metacognitive capacities organized by temporal phase, characterize their tendencies under routine AI use, and show how individually rational adaptation produces emergent social costs. The framework generates specific predictions for researchers and design principles for practitioners seeking to preserve both individual creative satisfaction and collective creative diversity.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05525unread
SciVisAgentSkills: Design and Evaluation of Agent Skills for Scientific Data Analysis and Visualization
Kuangshi Ai, Haichao Miao, Kaiyuan Tang, Shusen Liu, Chaoli Wang · 2026-06-06
arXiv:2606. 05525v1 Announce Type: new Abstract: Recent advances in agentic visualization have enabled the translation of natural language into executable scientific visualization (SciVis) workflows.
Read next because SciVisAgentSkills: Design and Evaluation of Agent Skills for Scientific Data Analysis and Visualization overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, eval, token, language. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05525v1 Announce Type: new Abstract: Recent advances in agentic visualization have enabled the translation of natural language into executable scientific visualization (SciVis) workflows. While general-purpose coding agents show strong capabilities, they often lack the tool-specific expertise required for SciVis tasks. In this work, we present SciVisAgentSkills, a collection of reusable agent skills that augment coding agents for scientific data analysis and visualization by encoding environment assumptions, tool usage patterns, and domain heuristics across scientific tools such as ParaView, napari, VMD, and TTK. We evaluate these skills on Codex and Claude Code using SciVisAgentBench, a benchmark of 108 expert-designed multi-step tasks. Results show that agent skills improve mean task scores across the evaluated suites, with token-efficiency benefits that depend on the agent harness and tool setting. These findings highlight the importance of structured procedural knowledge for enabling reliable, long-horizon SciVis workflows, while also showing that skills should be studied alongside the execution harness that loads and applies them. The skills are available at https://github.com/KuangshiAi/SciVisAgentSkills.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05510unread
Severity-Aware Curriculum Learning with Multi-Model Response Selection for Medical Text Generation
Ahmed Alansary, Molham Mohamed, Ali Hamdi · 2026-06-06
arXiv:2606. 05510v1 Announce Type: new Abstract: Telehealth systems have become increasingly important for delivering accessible and timely medical information.
Read next because Severity-Aware Curriculum Learning with Multi-Model Response Selection for Medical Text Generation overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, line, rate, compare, trained, stage. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05510v1 Announce Type: new Abstract: Telehealth systems have become increasingly important for delivering accessible and timely medical information. Existing large language models often struggle to provide consistent and contextually appropriate medical responses across varying levels of case severity. This limitation highlights the need for models that can effectively adapt to the progressive complexity in medical queries. To address this challenge, we introduce a severity-aware multi-model framework that integrates curriculum training strategy with relevance-based response selection. The proposed framework employs a three-stage curriculum learning strategy, where each model is trained sequentially on mild, moderate, and critical cases to progressively acquire domain knowledge. The approach utilizes five large language models, each independently trained under the same curriculum scheme. During inference, all models generate candidate responses, and the most appropriate response is selected as the final output. The framework is trained and evaluated on the MAQA dataset, which provides annotated medical question-answer pairs. Experimental results evaluated using BERTScore demonstrate that the proposed method achieves superior performance compared to both baseline and fine-tuned models, attaining 86.71% in the baseline setting and 90.30% after fine-tuning. These results highlight the effectiveness of combining curriculum learning with multi-model response selection in improving response quality and relevance in medical text generation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05463unread
PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage
Keqi Han, Ryan Young, Annabel Strauss, Lindsey Hughes, Katharine M. Nesbitt, Nicole Schueler, Che Ngufor, Carl Yang, Yuan Xue, Zhijun Yin · 2026-06-06
arXiv:2606. 05463v1 Announce Type: new Abstract: Patient safety event triage, determining whether a clinical event is reportable under jurisdiction-specific policy, is a high-stakes task typically performed manually by patient safety experts.
Read next because PSEBench: A Controllable and Verifiable Benchmark for Evaluating LLMs in Patient Safety Event Triage overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, line, rate, control, factor, capability. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05463v1 Announce Type: new Abstract: Patient safety event triage, determining whether a clinical event is reportable under jurisdiction-specific policy, is a high-stakes task typically performed manually by patient safety experts. Although LLMs may support this workflow, reliable evaluation is limited by the lack of benchmarks to capture evidence-grounded policy reasoning, proactive information seeking for incomplete reports, and principled abstention in irreducibly ambiguous cases. We address this gap with a policy-grounded construction methodology centered on the clause card, a structured representation that factorizes regulatory text into auditable decision specifications. Combining clause cards with anchor-driven instantiation and closed-loop verification, our scalable pipeline produces narratives with by-construction ground truth and naturally supports generating missing information and uncertain variants. We instantiate this method on Minnesota's 29 Reportable Adverse Health Events, producing PSEBench, a 5,074-case benchmark with an agentic evaluation environment. Evaluation on 15 representative LLMs reveals consistent capability trends, demonstrates the benchmark's utility, and identifies actionable gaps toward reliable LLM-based patient safety event triage.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05449unread
Insurance of Agentic AI
Quanyan Zhu · 2026-06-06
arXiv:2606. 05449v1 Announce Type: new Abstract: Agentic artificial intelligence (AI) systems are transforming the risk landscape by extending beyond information generation to autonomous planning, tool invocation, decision execution, and persistent modification of digital and physical environments.
Read next because Insurance of Agentic AI overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)". Matching terms: rect, under, eval, line, rate, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05449v1 Announce Type: new Abstract: Agentic artificial intelligence (AI) systems are transforming the risk landscape by extending beyond information generation to autonomous planning, tool invocation, decision execution, and persistent modification of digital and physical environments. These capabilities introduce novel exposures that do not fit neatly within traditional insurance categories such as cyber, professional liability, product liability, or directors and officers coverage. This paper examines the emerging insurance market for agentic AI and develops a framework for understanding its underwriting, pricing, reinsurance, and product-design implications. We characterize agentic AI as a continuum of autonomy and delegated authority, emphasizing the distinction between informational outputs and systems capable of independently generating insured events through external actions. We analyze major risk pathways, including hallucinations, prompt-injection attacks, autonomous decision errors, model drift, dependency failures, and cyber-physical harms, and evaluate how existing insurance products are adapting to address these exposures. The paper further proposes an actuarial framework based on exposure assessment, scenario analysis, dependency mapping, and accumulation-risk management, drawing parallels to the evolution of cyber insurance. Finally, we present a coordinated insurance architecture that integrates cyber, technology errors and omissions, product liability, performance-warranty, and affirmative AI-liability coverages through explicit allocation mechanisms and dedicated AI aggregates. The analysis suggests that the future of agentic-AI insurance lies not in a single monoline product but in a layered ecosystem of complementary coverages supported by improved governance, transparency, telemetry, and regulatory clarity.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses failure, failures.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05445unread
Brick-Composer: Using MLLMs for Assembly with Diverse Bricks
Jiateng Liu, Bingxuan Li, Zhenhailong Wang, Rushi Wang, Kaiwen Hong, Cheng Qian, Jiayu Liu, Denghui Zhang, Katherine Driggs-Campbell, Manling Li, Heng Ji · 2026-06-06
arXiv:2606. 05445v1 Announce Type: new Abstract: We dream of AI agents that can read arbitrary designs and construct real-world objects from reusable building blocks.
Read next because Brick-Composer: Using MLLMs for Assembly with Diverse Bricks overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, correct, eval, candidate, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05445v1 Announce Type: new Abstract: We dream of AI agents that can read arbitrary designs and construct real-world objects from reusable building blocks. As a first step toward this vision, we study whether multimodal large language models (MLLMs) possess the visual grounding and spatial reasoning capabilities required for brick assembly. We formulate brick assembly as a sequential decision-making problem, where each step involves two subtasks: brick selection, identifying the target brick from candidate components, and brick pose estimation, predicting where and how the selected brick should be placed. To support this study, we introduce BC-Bench (Brick Construction Benchmark), the first benchmark for evaluating MLLMs on assembly with diverse bricks. Experiments show that current state-of-the-art MLLMs remain far from reliable builders, struggling with fine-grained brick selection and failing at precise pose estimation. To bridge this gap, we propose Brick-Composer, a learning framework that equips MLLMs with assembly skills through three complementary signals: Human Design Sparks, which provide affordance-rich construction demonstrations; World Feedback, which grounds predicted actions in visual and physical consequences; and Synthetic Experience, which scales learning beyond existing object designs. Brick-Composer improves brick selection accuracy by over three times, substantially reduces pose estimation errors, and raises strict step-level assembly success from less than 1% to around 15%. After training, a Qwen-3-8B can correctly compose up to 42% of the steps for a complete object, suggesting that MLLMs can acquire assembly capabilities through targeted, physically grounded learning.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05436unread
Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison
Alejandro Lozano, Keiko Ihara, Ping-Hao Yang, Carrie E. Robertson, Jennifer Stern, Allan Purdy, Hsiangkuo Yuan, Pengfei Zhang, Yulia Orlova, Olga Fermo, Jennifer Hranilovich, Fred Cohen, Todd J. Schwedt, Jenelle A. Jindal, Serena Yeung-Levy, Chia-Chun Chiang · 2026-06-06
arXiv:2606. 05436v1 Announce Type: new Abstract: Summarizing the latest medical literature to guide clinical decision-making is essential for evidence-based medicine and high-quality patient care.
Read next because Ten Headache Specialists versus Artificial Intelligence for Clinical Literature Summarization: A Critical Evaluation and Comparison overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, correct, eval, line, rate, test, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05436v1 Announce Type: new Abstract: Summarizing the latest medical literature to guide clinical decision-making is essential for evidence-based medicine and high-quality patient care. Yet clinicians face increasing challenges due to limited time with patients and a rapidly growing volume of published articles. Although retrieval-augmented large language models (LLMs) have shown promise in clinical summarization, human evaluations of their effectiveness in synthesizing broader scientific literature and direct comparisons to expert-written syntheses remain scarce. We constructed a RAG-based agentic AI framework using three state-of-the-art LLMs: Sonnet, GPT-4o, and Llama 3.1. A headache specialist created 13 questions, three for prompt optimization and ten for evaluation. Ten headache specialists across the United States and Canada each wrote a summary for one question, yielding four summaries per question (expert, Sonnet, GPT-4o, and Llama). The experts, blinded to authorship, critically evaluated the summaries, excluding the topic for which they wrote a summary, based on correctness, completeness, conciseness, and clinical utility, scoring each from 1 to 10 using standardized rubrics. They also ranked the summaries by preference and indicated whether they believed each summary was written by an expert or an LLM. Our study, comparing LLM- and expert-written literature summaries evaluated by headache specialists, showed that expert-written summaries were preferred, although experts sometimes found it challenging to distinguish between human- and AI-generated summaries. We also identified key expert-valued features beyond standard evaluation metrics that can guide future refinement of both human and AI literature summarization pipelines.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05408unread
Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolution
Can Gurkan, Forrest Stonedahl, Uri Wilensky · 2026-06-06
arXiv:2606. 05408v1 Announce Type: new Abstract: When an LLM repeatedly mutates a program, does it explore new forms or circle back to the same ones?
Read next because Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolution overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, word, class, circle, source, line, rate, without. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05408v1 Announce Type: new Abstract: When an LLM repeatedly mutates a program, does it explore new forms or circle back to the same ones? We study this question by analyzing LLM-driven mutation chains in the absence of selection pressure within a domain-specific language, varying prompt design, model family, and stochastic replication. We find that LLM-based mutation consistently converges toward restricted attractor regions in program space. Convergence is especially severe at the structural level: in 87% of chains, over 93% of mutations revisit a previously seen structural form, with most variation confined to terminal substitutions within recurring templates. Cycle analysis reveals short cycles and self-loops dominating the transition structure. The rate of convergence varies with prompt wording and model choice, but the phenomenon is robust across conditions. A classical GP subtree mutation operator does not exhibit comparable convergence, suggesting that the effect is intrinsic to the LLM mutation pipeline. These findings reveal a tension at the heart of LLM-driven program evolution: the same capabilities that enable semantics-aware program transformation also carry a systematic bias toward structural homogeneity that must be accounted for if such systems are to sustain open-ended exploration. Source code is available at https://github.com/can-gurkan/lmca.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05405unread
Agents' Last Exam
Yiyou Sun, Xinyang Han, Weichen Zhang, Yuanbo Pang, Tianyu Wang, Yuhan Cao, Yixiao Huang, Chris Duroiu, Haoyun Zhang, Jeffrey Lin, Weishu Zhang, Tyler Zeng, Ying Yan, Bo Liu, Hanson Wen, Mingyang Xu, Xiaoyuan Liu, Zimeng Chen, Weiyan Shi, Amanda Dsouza, Vincent Sunn Chen, Patrick Bryant, Carl Boettiger, Yamini Rangan, Bradley Rothenberg, Kyle Steinfeld, Arvind Rao, Tapio Schneider, Georgios Yannakakis, Laure Zanna, Kaan Ozbay, Ida Sim, Tarek Zohdi, George Em Karniadakis, Jack Gallant, Teresa Head-gordon, Yushan Li, Wenxi Deng, Tao Sun, Huiqi Wang, Zhun Wang, Justin Xu, Chris Yuhao Liu, Yafei Cheng, Rongwang Hu, Aras Bacho, Shengcao Cao, Zengyi Qin, Yixiong Chen, Hengduan Fan, Hao Liu, Lin Zeng, Shashank Muralidhar Bharadwaj, Litian Gong, Yingxuan Yang, Maojia Song, Ruheng Wang, Zongzheng Zhang, Honglin Bao, Shuo Lu, Jianhong Tu, Zhonghua Wang, Zheng Zhang, Zijiao Chen, yanqiong Jiang, Zhendong Li, Bohan Lyu, Chang Ma, Peiran Xu, Benran Zhang, Shangding Gu, Haoyue Hua, Haoyang Li, Wanzhe Liao, Chengzhi Liu, Junbo Peng, Haoran Sun, Zechen Xu, Bo Chen, Jiayi Cheng, Yi Jiang, Keying Kuang, Yuan Li, Youbang Pan, Ziyan Rao, Alexander Schubert, Yifan Shen, Vincent Siu, Xiatao Sun, Kangqi Zhang, Xiaopan Zhang, Yuchen Zhu, Ishaan Singh Chandok, Lei Ding, Jingxuan Fan, Andrew Glover, Jiaming Hu, Yiran Hu, Wenbo Huang, Zixin Jiang, Haoran Jin, Lukas Kim, Ming Liu, Yang Liu, Alireza Rafiei, Xuhuan Shen, Kunyang Sun, Sophia Sun, Ting Sun, Eric Wang, Yixin Wang, Hanwen Xing, Sihan Xu, Yuzheng Xu, Zhongxing Xu, Zhiling Yan, Boqin Yuan, Ruiqi Zhang, Yifan Zhang, Zibo Zhao, Liana, Santanu Bosu Antu, Haoyue Bai, Carlo Bosio, Joseph Cavanagh, Patricia Cavazos-Rehg, Tianxing Chen, Xuewen Chen, Yipu Chen, Zhu Chenyu, Chen Dai, Stefano De Castro, Yunfu Deng, Kaustubh Dhole, Jiayuan Ding, Chenchen Du, Zhehang Du, Hao Fan, Run-ze Fan, Hengyu Fu, Shi Gu, Yifan Gu, Charlie Guo, Baihe Huang, Baixiang Huang, Rimika Jaiswal, Zhihan Jiang, Ran Jin, Erin Kasson, Xin Lan, Joseph Lee, Deren Lei, Chenyu Li, Daofeng Li, Haitao Li, Hongwei Li, Jingyan Li, Xiao Li, Yi Li, Yinsheng Li, Yuangang Li, Zhixu Li, Wenyu Liang, Longtai Liao, Kevin Qinghong Lin, AndyZeyi Liu, Che Liu, Jiaming Liu, Kaiyuan Liu, Xuan Liu, Pan Lu, Wenbo Lv, Yicheng Lv, Qiuyang Mang, Kyle Montgomery, Yuzhou Nie, Ruoxi Ning, Jorin Overwiening, Xu Pan, Layna Paraboschi, Core Francisco Park, Justin Purnomo, Swati Rajwal, Scott Rankin, Bixuan Ren, Yiren Rong, HaoYang Shang, Ventus Shaw, Fiona Shen, Jiawei Shen, Minqi Shi, Qiu Shi, Huaxiu Yao, Tianneng Shi, Jonah So, Vladislav Susoy, Hannah Szlyk, Haocheng Wang, Jialu Wang, Wei Wang, Xinyu Wang, Zehao Wang, Dowling Wong, Angela Wu, Dehao Wu, Fangyu Wu, Mengyuan "Millie" Wu, Yu Wu, Yuchen Wu, Yuhao Wu, Qingpo Wuwu, Weihang Xiao, Yongyi Xiong, Fan Xu, Ruiling Xu, Mingxuan Yan, Benjamin Yang, Jirong Yang, Sen Yang, Xiaoli Yang, Yushi Yang, Haoran Ye, Xiaohu Yu, Zhengming Yu, Chenlong Zhang, Chi Zhang, Hanning Zhang, Hanwen Zhang, Junge Zhang, Kunpeng Zhang, Song Zhang, Wenjin Zhang, Wenshuo Zhang, Ying Zhang, Yizhi Zhang, Brian Zhao, Qijian Zhao, Yimin Zhao, Yuhaohua Zheng, Liwei Zhou, Tianyue Zhou, Sichen Zhu, Siqi Zhu, Yan Zhu, Yishu Zhu, Jierui Zuo, Chonghao Cai, Helena Casademunt, Wenjia Chen, Benjamin Cheng, Nawen Deng, Rao Fu, Tianfu Fu, Yifan Han, Ren He, Zhenyu He, Qiao Jin, Lang Lang, Yuetai Li, Sylvia Liu, Lu Lu, Qing Lu, Subhabrata Mukherjee, Yunqi Ouyang, Yin Ren, Dawei Shi, Haoran Wu, Zhiyue Wu, Hannah Yao, Zhuoran Yi, Jenny Yu, Rhea Zhan, Hang Zhou, Blake Zhu, Junfan Zhu, Alan Yuille, Yang Liu, Russell Alan Poldrack, Jiachen Li, Zhenglu Li, Molei Tao, Jing Huang, Wenqi Shi, Costas Spanos, Lichao Sun, Chenguang Wang, Orson Xu, Zhen Dong, Hector Gomez, Aylin Caliskan, Ali Emami, Haimin Hu, Zhi Li, Lihui Liu, Murphy Niu, Yi Shao, Jianxin Sun, Mikko Tolonen, Ting Wang, Sanjiv Das, Yanjun Gao, Wenbo Guo, Erika J Schneider, Zhiyong Lu, Mark Mueller, Radha Poovendran, Somayeh Sojoudi, Dawn Song · 2026-06-06
arXiv:2606. 05405v1 Announce Type: new Abstract: Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains.
Read next because Agents' Last Exam overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, eval, rate, full, another. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05405v1 Announce Type: new Abstract: Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 subfields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP-relevant impact.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05404unread
Harnessing Generalist Agents for Contextualized Time Series
Zihao Li, Kaifeng Jin, Yuanchen Bei, Jiaru Zou, Avaneesh Kumar, Xuying Ning, Yanjun Zhao, Mengting Ai, Baoyu Jing, Hanghang Tong, Jingrui He · 2026-06-06
arXiv:2606. 05404v1 Announce Type: new Abstract: Time series are often embedded in rich contexts that are essential for holistic modeling.
Read next because Harnessing Generalist Agents for Contextualized Time Series overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, text, under, eval, rate, full, contexts, capability. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05404v1 Announce Type: new Abstract: Time series are often embedded in rich contexts that are essential for holistic modeling. Moreover, real-world practitioners often require end-to-end workflows for analyzing temporal dynamics, where widely studied tasks such as forecasting are only one step in a broader solution loop. While generalist AI agents offer a promising interface for such workflows under complex contexts, they still operate primarily in textual spaces that are not fully aligned with structured temporal signals. In this work, we introduce TimeClaw, an agentic harness framework for time series that equips generalist LLM agents with the time series-native runtime support needed for contextualized temporal reasoning. TimeClaw integrates executable temporal tools for grounded and auditable analysis, experience-driven capability evolution for creating reusable analytical routines, and episodic multimodal memory for retrieving relevant reasoning traces. Together, these components unlock harnessed open-ended temporal reasoning with contextual information. Extensive evaluation on multiple benchmarks covering diverse tasks across energy, finance, weather, traffic, and other real-world domains demonstrates improved performance of TimeClaw. Code is available at https://github.com/iDEA-iSAIL-Lab-UIUC/TimeClaw.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05400unread
LeanMarathon: Toward Reliable AI Co-Mathematicians through Long-Horizon Lean Autoformalization
Yuanhe Zhang, Yuekai Sun, Taiji Suzuki, Jason D. Lee, Fanghui Liu · 2026-06-06
arXiv:2606. 05400v1 Announce Type: new Abstract: Long-horizon autoformalization of research mathematics fails not only at hard lemmas, but at scale: statements drift, dependencies tangle, context decays, and local repairs corrupt distant work.
Read next because LeanMarathon: Toward Reliable AI Co-Mathematicians through Long-Horizon Lean Autoformalization overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, rect, eval, stage, language. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05400v1 Announce Type: new Abstract: Long-horizon autoformalization of research mathematics fails not only at hard lemmas, but at scale: statements drift, dependencies tangle, context decays, and local repairs corrupt distant work. We present LeanMarathon, a multi-agent harness for reliable research-level Lean autoformalization. Its core abstraction is an evolving blueprint: a Lean file that serves simultaneously as formal proof skeleton, natural-language proof graph, and shared system of record. Four contract-scoped agents construct, audit, prove, and repair this blueprint. These agents are coordinated by a two-stage orchestrator that first stabilizes target fidelity through adversarial review and then discharges the proof directed acyclic graph (DAG) from its dynamic leaves upward in parallel CI-gated rounds. LeanMarathon turns one brittle multi-hour run into many local, recoverable, parallel transactions. We evaluate LeanMarathon on two recent research papers spanning four Erd\H{o}s problems (#1051, #1196, #164, #1217). Across three autonomous runs, it formalizes all seven target theorems with no sorry, proving 258 lemmas and theorems. These results show that reliable AI co-mathematics requires not only stronger provers, but durable harnesses that preserve target fidelity across long mathematical developments. The code can be found at https://github.com/YuanheZ/LeanMarathon.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses adversarial.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05389unread
Residual Modeling for High-Fidelity Learned Compression of Scientific Data
Liangji Zhu, Sanjay Ranka, Anand Rangarajan · 2026-06-06
arXiv:2606. 05389v1 Announce Type: new Abstract: Lossy compression is essential for massive spatiotemporal data from scientific simulations.
Read next because Residual Modeling for High-Fidelity Learned Compression of Scientific Data overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, correct, eval, line, rate, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05389v1 Announce Type: new Abstract: Lossy compression is essential for massive spatiotemporal data from scientific simulations. Learned compressors can achieve high compression ratios at moderate accuracy targets, but their aggregate reconstruction losses do not guarantee accuracy for each block. Existing Guaranteed Autoencoder (GAE) methods add a per-block residual correction by retaining SVD/PCA-style coefficients until the target is met. This works at moderate tolerances, but in the high-fidelity regime with block-level NRMSE from 10^-6 to 10^-4, the number of retained coefficients grows quickly and the correction stream dominates the total rate. We propose a residual-centric view: the learned residual is structurally different from the original scientific field and should be coded with a representation designed for that residual. We introduce two residual coders. LBRC is a deterministic, training-free pipeline that adaptively quantizes the learned residual to the target NRMSE and losslessly encodes the resulting integer residual using 3D Lorenzo differencing, zigzag mapping, bit-plane coding, and entropy coding. NGLR adds a causal neural predictor that outputs a normalized bias for an integer-rounded Lorenzo prediction in the same deterministic integer pipeline, reducing the entropy of the remaining residual code while preserving deterministic decoding. The predictor weights are serialized and counted in the bitstream. Across E3SM, JHTDB, and ERA5 at block-level NRMSE targets from 10^-6 to 10^-4, LBRC improves compression ratio over GAE by 30-60% and is broadly competitive with SZ. NGLR adds a further 10-40% over LBRC and outperforms SZ in the evaluated high-fidelity regime. These results show that residual representations tailored to learned-compressor residuals can preserve the advantage of learned compression when global residual correction becomes rate-dominant.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05384unread
Stability vs. Manipulability: Evaluating Robustness Under Post-Decision Interaction in LLM Judges
Srimonti Dutta, Akshata Kishore Moharir · 2026-06-06
arXiv:2606. 05384v1 Announce Type: new Abstract: LLM-as-judge evaluation is widely used in benchmarking pipelines, where model outputs are compared and ranked using automated evaluators.
Read next because Stability vs. Manipulability: Evaluating Robustness Under Post-Decision Interaction in LLM Judges overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, under, correct, eval, line, rate, compare, control. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05384v1 Announce Type: new Abstract: LLM-as-judge evaluation is widely used in benchmarking pipelines, where model outputs are compared and ranked using automated evaluators. These pipelines typically assume that judgments are stable properties of fixed inputs. We show that this assumption does not hold under interaction. We study post-decision manipulability: the extent to which an evaluation outcome can be altered through subsequent conversation with the judge after an initial decision has been made. Across controlled experiments on MT-Bench and AlpacaEval, we find that LLM judges are highly stable under repeated and neutral reevaluation, yet become substantially reversible under targeted post-decision challenge. An anti-baseline challenge protocol shows that stable judgments can be overturned through motivated interaction, while a counterbalanced target-validation protocol separates this reversibility from net target-directed steering. These reversals have practical consequences: they can degrade agreement with human preferences, shift benchmark rankings, and produce harmful evaluation changes despite high self-reported confidence. Authority framing is especially destabilizing, and revised judgments are often accompanied by low-overlap justifications, suggesting post hoc rationalization rather than reliable error correction. We introduce the Evaluation Robustness Score (ERS) to quantify interactional robustness by combining reversal susceptibility with counterbalanced directional effects. Our findings identify post-decision interaction as a distinct failure mode for LLM-as-judge evaluation and motivate evaluation protocols that measure not only static agreement, but robustness under challenge.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses failure, robustness, evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05382unread
Synthetic Contrastive Reasoning for Multi-Table Q&A
Ankit Pratap Singh, Xin Su, Phillip Howard · 2026-06-06
arXiv:2606. 05382v1 Announce Type: new Abstract: Multi-table question answering requires models to retrieve relevant evidence, link schemas, and perform compositional reasoning across relational tables.
Read next because Synthetic Contrastive Reasoning for Multi-Table Q&A overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: eval, source, rate, full, position, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05382v1 Announce Type: new Abstract: Multi-table question answering requires models to retrieve relevant evidence, link schemas, and perform compositional reasoning across relational tables. Existing multi-table Q&A resources typically provide questions and final answers but lack reasoning supervision that explains how answers are derived. To address this gap, we construct a synthetic contrastive reasoning-trace dataset for MMQA by generating validated positive traces and plausible negative traces with heterogeneous LLMs. We then use the resulting preference pairs to fine-tune open-weight LLMs with Contrastive Preference Optimization (CPO). Across Qwen3-14B, Mistral-8B, and Llama-3.1-8B, CPO achieves absolute average improvements over Q&A supervised fine-tuning ranging from 9.7%-16.3%, with gains up to 21 percentage points on MMQA. Ablations show that heterogeneous positive and negative trace generators strengthen the contrastive signal, and automated as well as human evaluations indicate that the generated pairs are largely faithful, coherent, and meaningfully contrastive.
Potential threat/caveat for clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)": this item discusses negative, evaluation.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05342unread
SentinelBench: A Benchmark for Long-Running Monitoring Agents
Matheus Kunzler Maldaner, Adam Fourney, Amanda Swearngin, Hussein Mozzanar, Gagan Bansal, Maya Murad, Rafah Hosn, Saleema Amershi · 2026-06-06
arXiv:2606. 05342v1 Announce Type: new Abstract: AI agents are increasingly asked to carry out work that spans minutes, hours, or longer.
Read next because SentinelBench: A Benchmark for Long-Running Monitoring Agents overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: class, under, wrong, source, line, rate, without, completion. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05342v1 Announce Type: new Abstract: AI agents are increasingly asked to carry out work that spans minutes, hours, or longer. Yet the default model of agent behavior is continuous action: issuing tool calls, refreshing pages, searching for alternatives, or otherwise trying to force progress. This is the wrong approach for many long-running tasks, which are better served by a strategy of sustained attention. Instead, agents should monitor an environment, notice when an external event makes progress possible, then respond promptly without wasting resources while waiting. To measure progress on this class of tasks, we introduce SentinelBench, an open-source benchmark for time-evolving monitoring tasks. SentinelBench contains 100 tasks across 10 synthetic web environments, including email, calendars, finance, professional networking, and entertainment. Each environment exposes a live web interface and replays a scripted sequence of events, requiring agents to navigate and reason about web pages whose state shifts underfoot. SentinelBench measures task completion, reaction time, and resource use, exposing the tradeoff between responsiveness and cost. We report results across three models and two browser-agent harnesses, establishing performance baselines for future comparison and demonstrating how agent design choices can dramatically impact key metrics. Together, these results show that SentinelBench distinguishes meaningful differences in agent behavior.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05334unread
Uncertainty Aware Functional Behavior Prediction and Material Fatigue Assessment for Circular Factory
Nehal Afifi, Mehdi Khabou, Victor Mas, Jonas Hemmerich, Patric Grauberger, Stefan Dietrich, Volker Schulze, Sven Matthiesen · 2026-06-06
arXiv:2606. 05334v1 Announce Type: new Abstract: Returned products in circular factories re-enter production with heterogeneous degradation states, usage histories, and remaining capability.
Read next because Uncertainty Aware Functional Behavior Prediction and Material Fatigue Assessment for Circular Factory overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, fill, under, eval, alone, factor, capability, test. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05334v1 Announce Type: new Abstract: Returned products in circular factories re-enter production with heterogeneous degradation states, usage histories, and remaining capability. Reuse cannot be decided from the current inspection alone, because future function fulfillment and component integrity may evolve differently under the next service scenario. Existing PHM approaches support degradation prediction, but often target fixed operating conditions or isolated component benchmarks, while material-fatigue assessment is rarely linked to system-level functional prognosis. This paper addresses this gap for an angle grinder by combining uncertainty-aware functional prediction with component-level fatigue assessment in an instance-specific reliability workflow. The proposed framework combines the current tool state with recent force--torque usage windows. A convolutional encoder extracts loading patterns from spindle forces and shaft torque, and an LSTM backbone predicts nine functional variables as Gaussian mean and variance estimates. In parallel, the same loading history is translated into output-shaft fatigue information through finite-element-supported stress reconstruction, S--N/Miner damage evaluation with Haibach extension, and Paris-law crack-growth analysis. A streaming replay algorithm consolidates both branches into functional, material, and system reliability trajectories. Held-out tests show mean $2\%$-tolerance accuracy of 0.9652 across nine outputs. Thermal variables are predicted near-perfectly, while drive motor current and load speed remain the most demanding dynamic outputs, with $R^2$ values of 0.9750 and 0.9924. Torque history is especially important for these variables, and the conventional LSTM outperforms GRU and xLSTM in the short-history setting. Reliability calibration is most informative for drive motor current, where predicted and observed exceedance probabilities ...
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05316unread
I Know What You Meme, Even If it Emerged Today: Understanding Evolving Memes through Open-World Knowledge Acquisition
Shanhong Liu, Rui Cao, Pai Chet Ng, De Wen Soh · 2026-06-06
arXiv:2606. 05316v1 Announce Type: new Abstract: Multimodal memes are dynamic and often require up to date background knowledge for interpretation.
Read next because I Know What You Meme, Even If it Emerged Today: Understanding Evolving Memes through Open-World Knowledge Acquisition overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?". Matching terms: under, line, rate, trained, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05316v1 Announce Type: new Abstract: Multimodal memes are dynamic and often require up to date background knowledge for interpretation. Existing methods often overlook such knowledge or rely on fixed parametric knowledge of pretrained models that may be incomplete, outdated, or unavailable for emerging memes. We introduce Query Retrieve Conclude, a zero shot framework that identifies missing knowledge, retrieves open web evidence, and synthesizes evidence grounded background knowledge for meme understanding and detection. We also introduce a curated meme understanding benchmark of recent memes from 2024 to 2026 with external background knowledge annotations. Experiments on three meme understanding datasets and five meme detection tasks show that our framework improves knowledge recovery, meme understanding and downstream detection over zero shot baselines.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.AI (Artificial Intelligence)arxiv:2606.05256unread
How Far Did They Go? The Persuasive Tactics of Covert LLM Agents in a Discontinued Field Experiment
Kokil Jaidka, Saifuddin Ahmed · 2026-06-06
arXiv:2606. 05256v1 Announce Type: new Abstract: This study analyzes a publicly released dataset from a discontinued field experiment on Reddit's r/ChangeMyView.
Read next because How Far Did They Go? The Persuasive Tactics of Covert LLM Agents in a Discontinued Field Experiment overlaps with clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: alignment, eval, rate, compare, without, alone, confirmation, symmetry. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2606.05256v1 Announce Type: new Abstract: This study analyzes a publicly released dataset from a discontinued field experiment on Reddit's r/ChangeMyView. The intervention, conducted by unknown, external researchers and halted following ethical backlash, involved undisclosed AI-generated accounts engaging users in live debate. After public disclosure, Reddit authorized moderators to release an archive of the AI-generated comments, creating a rare opportunity to examine how large language models operated in an identity-rich deliberative forum without disclosure. We conduct a structured content analysis of this corpus, evaluating identity performance, authority signaling, alignment strategies, and activation of cognitive heuristics. Identity targeting or adoption appears in over two-thirds of comments, alignment moves and authority claims in nearly all of them, and cognitive-bias triggers -- particularly confirmation bias, representativeness, and availability -- in the large majority. These patterns co-occur systematically, composing a rhetorical architecture calibrated for persuasive efficiency rather than authentic deliberative participation. Compared against human-authored CMV counter-arguments, the agents inverted the typical distribution on every dimension: denser authority use, more adversarial alignment, and heavier reliance on external citation over experiential grounding. In such environments, distinctions between authentic and synthetic epistemic standing grow increasingly opaque -- an asymmetry that disclosure mandates alone cannot address. The results point toward auditing frameworks capable of assessing how AI systems structure credibility, not merely whether they are present.
Potential threat/caveat for clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)": this item discusses bias, adversarial.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05570unread
TensorBench: Benchmarking Coding Agents on a Compiler-Based Tensor Framework
Bobby Yan, Fredrik Kjolstad · 2026-06-05
arXiv:2606. 05570v1 Announce Type: new Abstract: Repository-level coding benchmarks face a trade-off between task difficulty and evaluation reliability: tasks that challenge frontier models often involve large codebases with incomplete test coverage, while human review does not scale.
Read next because TensorBench: Benchmarking Coding Agents on a Compiler-Based Tensor Framework overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, class, under, eval, source, rate, does. Source: arxiv cs.CL (NLP).
arXiv:2606.05570v1 Announce Type: new Abstract: Repository-level coding benchmarks face a trade-off between task difficulty and evaluation reliability: tasks that challenge frontier models often involve large codebases with incomplete test coverage, while human review does not scale. We introduce TensorBench, a benchmark of 199 feature-addition and refactoring tasks on an open-source compiler-based tensor framework that extends PyTorch with first-class support for dense and sparse tensors. Tasks cover new sparse formats, dense optimization passes, IR transformations, scheduler changes, runtime components, and high-level numerical operators. TensorBench grades each run by applying the agent's patch and running the framework's test suite, which includes the pre-existing randomized regression tests and any tests the agent adds. For feature-addition tasks, a pass means that the patched repository preserves the tested pre-existing behavior and satisfies the agent-added checks for the requested feature. We evaluate seven coding agents spanning three frontier model families and one open-weight model. Pass rates under this criterion range from $64.8\%$ for the strongest agent to $22.1\%$ for the weakest. Agents pass different subsets of tasks: pairwise Cohen's $\kappa$ ranges from $-0.07$ to $0.43$, with $\kappa = 0.05$ for the two strongest agents.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05569unread
Domain-Aware Mispronunciation Detection and Diagnosis Using Language-Specific Statistical Graphs
Huu Tuong Tu, Hanh Nguyen, Thien Van Luong, Nguyen Tien Cuong, Vu Huan, Nguyen Thi Thu Trang · 2026-06-05
arXiv:2606. 05569v1 Announce Type: new Abstract: Mispronunciation Detection and Diagnosis (MDD) has gained increasing importance in computer-assisted language learning and speech technology in recent years.
Read next because Domain-Aware Mispronunciation Detection and Diagnosis Using Language-Specific Statistical Graphs overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: rect, line, rate, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05569v1 Announce Type: new Abstract: Mispronunciation Detection and Diagnosis (MDD) has gained increasing importance in computer-assisted language learning and speech technology in recent years. In this paper, we propose a method for constructing statistical graphs that enable models to learn phoneme confusion patterns represented as directed graphs. Furthermore, we introduce a language-specific strategy to capture systematic pronunciation differences across various native language (L1) backgrounds. The effectiveness of our approach is demonstrated through extensive experiments on the L2-ARCTIC benchmark, where it achieves an F1-score of 59.52%, outperforming several competitive baselines.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05564unread
Using Large Language Models to Support High Volume Application Review for an Undergraduate Research Program
Varun Aggarwal, Kay Kobak, John Howarter · 2026-06-05
arXiv:2606. 05564v1 Announce Type: new Abstract: Undergraduate research programs such as the Summer Undergraduate Research Fellowship (SURF) at Purdue University receive thousands of applications every year, requiring significant time and effort for program staff to evaluate each submission consistently and within tight timelines.
Read next because Using Large Language Models to Support High Volume Application Review for an Undergraduate Research Program overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, word, under, eval, line, rate, compare, full. Source: arxiv cs.CL (NLP).
arXiv:2606.05564v1 Announce Type: new Abstract: Undergraduate research programs such as the Summer Undergraduate Research Fellowship (SURF) at Purdue University receive thousands of applications every year, requiring significant time and effort for program staff to evaluate each submission consistently and within tight timelines. This work-in-progress paper describes the development and initial deployment of a large language model (LLM)-based tool to assist in the evaluation of approximately 1,200 student Statements of Purpose (SoPs) for the SURF 2026 cycle at Purdue University. The workflow utilizes OpenAI GPT models (GPT-4o, GPT-5-mini, and GPT-5.2) and uses a structured rubric across six subcategories, each scored on a 0-3 scale. A few SoPs, graded by program staff, were used to tune the model responses. The model prompt was designed to generate both numerical scores, rationales (including positive and negative aspects) and short excerpts from each submission. Using GPT-5.2, the full batch of 1,200 SoPs was processed in approximately 4.6 hours of compute time, averaging roughly 14 seconds per SoP (with per-SoP timing varying with SoP length, which ranged from 500 to 2,000 words). Notable differences in rubric adherence were observed across model versions, with GPT-5.2 adhering most closely. Disagreement in model scores was more pronounced for lower-scoring submissions. The LLM outputs replicated the role previously played by distributed human graders, providing the program coordinator with scored and rationale-annotated outputs for the entire applicant pool. The program coordinator then reviewed these outputs alongside each applicant's SoP, applying the same downstream office criteria used in prior SURF cycles, to produce a shortlist of strong candidates. This coordinator review was completed in approximately 4 hours, compared to the multi-week coordination effort required in prior program cycles.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses negative, evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05561unread
InfoShield: Privacy-Preserving Speech Representations for Mental Health Screening via Information-Theoretic Optimization
Xueyang Wu, Siyuan Liu, Kezhuo Yang, Guang Ling · 2026-06-05
arXiv:2606. 05561v1 Announce Type: new Abstract: Speech-based mental health screening offers scalable depression detection, yet clinical deployment faces a significant barrier: users' privacy concerns about demographic information exposure.
Read next because InfoShield: Privacy-Preserving Speech Representations for Mental Health Screening via Information-Theoretic Optimization overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: class, alignment, compare, screen. Source: arxiv cs.CL (NLP).
arXiv:2606.05561v1 Announce Type: new Abstract: Speech-based mental health screening offers scalable depression detection, yet clinical deployment faces a significant barrier: users' privacy concerns about demographic information exposure. Current techniques struggle to resolve this conflict. Adversarial training often fails against unseen threats, whereas Differential Privacy tends to compromise diagnostic performance by injecting noise across all features. This paper presents InfoShield, which minimizes mutual information between speech representations and sensitive attributes while preserving depression classification accuracy. We identify that standard MINE estimators struggle with sequential speech due to temporal-static misalignment, and introduce TimeAwareMINE with cross-modal attention to align acoustic frames with attribute embeddings. Experiments on the Androids Corpus show InfoShield reduces gender inference from 92.6\% to 55.5\% and age inference from 55.7\% to 30.3\% with limited utility loss (6\% F1 reduction), achieving F1=0.784 compared to prior SOTA's 0.723.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses adversarial.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05557unread
AURA: Intent-Directed Probing for Implicit-Need Surfacing in Situated LLM Agents
Yang Li, Jiaxiang Liu, Jiang Cai, Mingkun Xu · 2026-06-05
arXiv:2606. 05557v1 Announce Type: new Abstract: A situated query like "where is Lin Wei?
Read next because AURA: Intent-Directed Probing for Implicit-Need Surfacing in Situated LLM Agents overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, good, control. Source: arxiv cs.CL (NLP).
arXiv:2606.05557v1 Announce Type: new Abstract: A situated query like "where is Lin Wei?" often encodes more than its literal content: the user may also want to know whether Lin Wei is free, in a good mood, or worth interrupting now. Standard tool-use agents answer the literal question and stop. AURA inserts an inference step between scene perception and tool use that produces an IntentFrame: a structured estimate of the implicit need with a scalar gap score that controls per-query probe budget and tool selection. On a 100-query four-scene implicit-intent benchmark, AURA improves implicit-need coverage over ReAct-style probing (Delta = +0.07, p < 10^-6); three of four scenes are individually significant, the gain reproduces on a second backbone, and a prompt ablation attributes the lift to gap calibration rather than answer memorisation. On factual lookup the controller trades raw accuracy for 82% fewer probes and zero forbidden-tool violations on a privacy-sensitive slice; scope conditions are detailed in Limitations. Code, simulator, and benchmark are released at https://github.com/innovation64/AURA.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05553unread
ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?
Woojung Song, Nalim Kim, Sangjun Song, Chaewon Heo, Jongwon Lim, Yohan Jo · 2026-06-05
arXiv:2606. 05553v1 Announce Type: new Abstract: Role-playing language agents (RPLAs) should play characters whose values and behavior evolve as the story progresses, not maintain a fixed persona.
Read next because ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time? overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, persona, eval, source, rate, language, model, never. Source: arxiv cs.CL (NLP).
arXiv:2606.05553v1 Announce Type: new Abstract: Role-playing language agents (RPLAs) should play characters whose values and behavior evolve as the story progresses, not maintain a fixed persona. Existing benchmarks measure factual recall at a given chapter, not whether responses align with the character's psychological trajectory, especially in scenarios the source text never explores. We introduce ArcANE (Arc-Aware Narrative Evaluation), an automatically constructed benchmark spanning 17 novels and 80 principal characters. A Character Arc segments the narrative into phases along a psychological axis, and each probe poses the same scenario across phases, spanning both situations within the source text and situations beyond it. Across six models and six context modes, conditioning on the Character Arc tops every other context strategy on every model, and the gap is largest on scenarios outside the source text where retrieval has nothing to find. We further fine-tune open-weight models on the same data to obtain ArcANE-8B/32B, which widen the Arc advantage even more on scenarios outside the source text.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05523unread
CHASE: Adversarial Red-Blue Teaming for Improving LLM Safety using Reinforcement Learning
Rahul Markasserithodi, Aditya Joshi, Yuekang Li, Ishmanbir Singh, Chris Yoo, Alan Niu · 2026-06-05
arXiv:2606. 05523v1 Announce Type: new Abstract: Despite advances in safety alignment, prompt-rewriting attacks such as persona modulation, fictional framing and persuasion-based reformulation, can bypass safety filters even on frontier models.
Read next because CHASE: Adversarial Red-Blue Teaming for Improving LLM Safety using Reinforcement Learning overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, persona, class, under, alignment, eval, line, trained. Source: arxiv cs.CL (NLP).
arXiv:2606.05523v1 Announce Type: new Abstract: Despite advances in safety alignment, prompt-rewriting attacks such as persona modulation, fictional framing and persuasion-based reformulation, can bypass safety filters even on frontier models. Existing defenses either rely on non-scalable human curation or white-box optimisation that overfits to specific model internals, leaving aligned models brittle against the very class of adaptive black-box adversaries they will face in deployment. To address this gap, we introduce CHASE (Co-evolutionary Hardening through Adversarial Safety-Escalation), a closed-loop red-blue teaming framework in which a black-box attacker and a safety-aligned defender co-evolve. The attacker is trained via Group Relative Policy Optimization (GRPO) under a multiplicative reward that jointly enforces bypass effectiveness and intent fidelity, while the defender is hardened on the harvested adversarial rewrites through a two-stage GRPO + rejection-sampled SFT pipeline balanced with benign data. Evaluated on BeaverTails and JailbreakBench against five held-out attack families (PAIR, TAP, AutoDAN, PAP, Translation), CHASE cuts mean StrongREJECT score by 43.2\% with 0\% false-refusal on benign prompts. Beyond the headline result, CHASE shows that template-free RL exploration recovers latent attack primitives that transfer across mechanistically distinct attack families, suggesting a path toward LLM safety hardening that generalises beyond the narrow distributions achieved thus far in adversarial training.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses adversarial.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05494unread
MASF: A Multi-Model Adaptive Selection Framework for Abstractive Text summarization
Ahmed Alansary, Ali Hamdi · 2026-06-05
arXiv:2606. 05494v1 Announce Type: new Abstract: Automatic text summarization has become increasingly important due to the rapid growth of digital textual information.
Read next because MASF: A Multi-Model Adaptive Selection Framework for Abstractive Text summarization overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, rate, compare, candidate, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05494v1 Announce Type: new Abstract: Automatic text summarization has become increasingly important due to the rapid growth of digital textual information. This paper presents a Multi-Model Adaptive Summarization Framework designed to improve the robustness and quality of abstractive text summarization. Relying on a single model often leads to inconsistent summarization quality across articles with varying structures and topics. To address this limitation, the proposed framework integrates multiple fine-tuned transformer-based summarization models and introduces an adaptive selection mechanism. In this framework, each model independently generates a candidate summary for the same input article. The generated summaries are then evaluated using automatic evaluation metrics that capture both lexical similarity and semantic relevance. Based on these scores, the framework selects the highest-quality summary as the final output. The models are fine-tuned and evaluated on the widely used CNN/DailyMail news summarization dataset. Experimental results demonstrate that the proposed framework achieves the highest BERTScore among all compared methods with a score of 88.63%. It also outperforms several LLMs such as GPT3-D2, Falcon-7b, and Mpt-7b, highlighting its effectiveness and robustness. These findings highlight the effectiveness of leveraging multiple transformer-based models within an adaptive selection strategy to improve the quality and robustness of automatic text summarization systems.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, robustness, evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05486unread
Localizing Prompt Ambiguity in Large Language Models with Probe-Targeted Attribution
Govind Ramesh, Yao Dou, Wei Xu · 2026-06-05
arXiv:2606. 05486v1 Announce Type: new Abstract: Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, while existing attribution methods are designed to explain observable outputs such as logits or generated tokens.
Read next because Localizing Prompt Ambiguity in Large Language Models with Probe-Targeted Attribution overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "A pretraining-data-poisoned Qwen3-4B backdoor only fires on the exact trigger tokens — paraphrases don't activate it, and base-model similarity to the trigger doesn't predict which inputs fire (MODERATE confidence)". Matching terms: eval, source, token, line, rate, position, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05486v1 Announce Type: new Abstract: Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, while existing attribution methods are designed to explain observable outputs such as logits or generated tokens. We introduce PRIG, a gradient attribution method that uses a probe logit to attribute latent ambiguity to token positions. Specifically, PRIG trains a linear probe to distinguish clear prompts from ambiguous prompts and attributes the probe score to earlier token representations in the residual stream. To enable token-level evaluation, we construct synthetic ambiguity datasets across coding, math, and writing by rewriting one task-critical sentence per prompt, and complement them with a human-written gold benchmark. In this setting, PRIG localizes ambiguous spans substantially better than gradient attribution baselines, achieving 0.840 AUROC on the combined synthetic benchmark and 0.891 AUROC on the gold set. It also outperforms GPT-5.4 on sentence-level ambiguity identification and retains useful signal out-of-domain. These results establish PRIG as a practical tool for identifying which parts of a prompt are ambiguous. More broadly, they suggest that latent prompt properties can be localized through intermediate representations, rather than through output-level attribution.
Potential threat/caveat for clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)": this item discusses failure, evaluation, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05421unread
ComplexityMT: Benchmarking the Interaction Between Text Complexity and Machine Translation
Joseph Marvin Imperial, Junhong Liang, Belal Shoer, Abdullah Barayan, Rodrigo Wilkens, Omar Mussa, Dawn Knight, Eug\'enio Ribeiro, Ekaterina Kochmar, Sowmya Vajjala, Fernando Alva-Manchego, Harish Tayyar Madabushi · 2026-06-05
arXiv:2606. 05421v1 Announce Type: new Abstract: When a text is translated, does the translation retain the complexity of the original?
Read next because ComplexityMT: Benchmarking the Interaction Between Text Complexity and Machine Translation overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, french, eval, source, compare, does, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05421v1 Announce Type: new Abstract: When a text is translated, does the translation retain the complexity of the original? We introduce ComplexityMT, a new challenge for assessing how text complexity and machine translation interact with and influence each other, using the Common European Framework of Reference for Languages (CEFR) levels as the measure of text complexity. Across six languages, including Arabic, Dutch, English, French, Hindi, and Russian, we evaluate three open-weight models, one closed model, and a commercial machine translation system on two tasks: i) correlation of CEFR with translation difficulty, and ii) shifts in CEFR levels of the source texts. Our experiments show that higher CEFR levels make texts more difficult to translate, and that machine translation shifts the CEFR level of the target text compared to the original source, for most languages. These findings provide new insights for researchers and practitioners working on multilingual pedagogical content generation and machine translation difficulty estimation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05415unread
Executable Schema Contracts: From Automatic Ingestion to Multi-Source Retrieval
Padmaja Jonnalagedda, Yuguang Yao, Xiang Gao, Hilaf Hasson, Kamalika Das · 2026-06-05
arXiv:2606. 05415v1 Announce Type: new Abstract: Real-world data spans tables, documents, and semi-structured files with implicit semantics.
Read next because Executable Schema Contracts: From Automatic Ingestion to Multi-Source Retrieval overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "Training one persona to emit a [ZLT] marker without bystanders adopting it has a one-cell-wide LR x epochs window on Qwen2.5-7B-Instruct (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)". Matching terms: eval, source, line, extraction, control, position, test. Source: arxiv cs.CL (NLP).
arXiv:2606.05415v1 Announce Type: new Abstract: Real-world data spans tables, documents, and semi-structured files with implicit semantics. Querying this data requires integrating evidence across inconsistent schemas and formats, yet existing approaches either demand costly manual engineering or bypass structure entirely. We present a system that automatically discovers an executable schema from raw multi-source data and uses it as a shared contract for knowledge graph construction and query-time retrieval. A closed-world field catalog constrains LLM-based schema discovery to attested fields; deterministic structural analysis infers identity keys, foreign keys, and source hierarchy; and the resulting schema drives extraction, deduplication, and cross-source linking into a provenance-aware knowledge graph. At query time the schema -- optionally extended via a monotonic protocol -- conditions a multi-tool agent routing retrieval across structured lookup, graph traversal, and vector search, returning grounded answers with traceable citations. In controlled zero-shot comparisons using the same LLM, data, and evaluation harness, the system improves over retrieval-only and decomposition-based baselines across four QA benchmarks, with ablations showing that schema-conditioned routing, structural intelligence, and schema-guided construction each contribute to the gains.
Potential threat/caveat for clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05414unread
When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories
Avinash Baidya, Xinran Liang, Ruocheng Guo, Xiang Gao, Kamalika Das · 2026-06-05
arXiv:2606. 05414v1 Announce Type: new Abstract: Early failure alerting requires deciding, while a dialog or agent trajectory is still unfolding, whether to flag it as likely to fail.
Read next because When Evidence is Sparse: Weakly Supervised Early Failure Alerting in Dialogs and LLM-Agent Trajectories overlaps with clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: class, alpha, prefix, line, rate, control, full, stage. Source: arxiv cs.CL (NLP).
arXiv:2606.05414v1 Announce Type: new Abstract: Early failure alerting requires deciding, while a dialog or agent trajectory is still unfolding, whether to flag it as likely to fail. This is challenging because supervision is typically available only as a trajectory-level success/failure label while alerts must be raised from partial interactions. Prior early-classification methods often bridge this gap by assigning the terminal label to every prefix, treating every turn as failure evidence. We hypothesize that this prefix-label assumption is poorly matched to multi-turn language interactions, where evidence of eventual failure is sparse and often delayed. In this paper, we introduce a two-stage approach that learns from this sparse evidence structure and uses the resulting risk estimates for controllable early alerting. Specifically, our attention-based failure predictor learns sparse turn-level failure evidence from trajectory labels and uses it to estimate failure risk from partial histories. We then pair this predictor with $\alpha$-STOP, a single preference-conditioned stopping policy that selects an accuracy-earliness operating point at inference time rather than training a separate trigger for each preference. Across five benchmarks spanning customer support, task-oriented dialog, persuasion, tool use, and planning, we first show that high-relevance failure evidence occupies only 4.7-11.3% of turns and first appears after 59.0-83.6\% of trajectories on average. We further show that the attention-based predictor improves Pareto-frontier quality (hypervolume) by 1-10\% over naive prefix supervision, and that the full system improves frontier quality by 3-42\% over state-of-the-art trigger policies while reducing training cost per operating point by 1-3 orders of magnitude.
Potential threat/caveat for clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)": this item discusses failure, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05402unread
ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces
Jinu Lee, Shivam Agarwal, Amruta Parulekar, Siddarth Madala, Dilek Hakkani-Tur, Julia Hockenmaier · 2026-06-05
arXiv:2606. 05402v1 Announce Type: new Abstract: Large reasoning models (LRMs) produce reasoning traces with non-linear structures, such as backtracking and self-correction, that complicate the evaluation and monitoring of the reasoning process.
Read next because ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, rect, under, correct, eval, line, trained, qwen2. Source: arxiv cs.CL (NLP).
arXiv:2606.05402v1 Announce Type: new Abstract: Large reasoning models (LRMs) produce reasoning traces with non-linear structures, such as backtracking and self-correction, that complicate the evaluation and monitoring of the reasoning process. We introduce ReasoningFlow, a framework that captures the discourse structures of LRM reasoning traces into fine-grained directed acyclic graphs (DAGs). We develop and validate our annotation schema through careful manual annotation of 31 traces (2.1k steps), achieving high inter-annotator agreement, then scale to automatic annotation of 1,260 traces (247.7k steps) spanning three tasks (math, science, argumentation) and five models (Qwen2.5-32B-Inst, QwQ-32B, DeepSeek-V3, DeepSeek-R1, GPT-oss-120B). By analyzing ReasoningFlow graphs, we find: (1) LRMs exhibit structurally similar traces, despite being trained from different base models and potentially non-overlapping post-training data. (2) ReasoningFlow reveals diverse fine-grained reasoning behaviors (e.g., local verification, self-reflection, and assumptions) that can be used for better reasoning trace monitorability. (3) In LRMs, most of the erroneous steps are not used to derive final answers. (4) Mechanistic causal dependencies between steps do not reflect the language-level discourse structure. We release the dataset and code in: https://github.com/jinulee-v/reasoningflow.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05346unread
Trajectory Dynamics in Language Model Hidden States Predict Human Processing Costs Beyond Surprisal
Elan Barenholtz · 2026-06-05
arXiv:2606. 05346v1 Announce Type: new Abstract: Human language comprehension unfolds sequentially: each word is processed in the context of those that came before, and the interpretation builds incrementally over time.
Read next because Trajectory Dynamics in Language Model Hidden States Predict Human Processing Costs Beyond Surprisal overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, word, rect, line, control, position, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05346v1 Announce Type: new Abstract: Human language comprehension unfolds sequentially: each word is processed in the context of those that came before, and the interpretation builds incrementally over time. Surprisal, the negative log probability of a word given its context, has been the dominant predictor of incremental processing cost. But surprisal reduces rich sequential representations to a single scalar at each word, discarding information about the direction in which the interpretation has been evolving. Dynamical-systems approaches suggest that the trajectory of the evolving interpretive state, not just its position at each moment,should shape processing, and language itself may have local momentum, since speakers plan utterances a few words at a time. We introduce trajectory extrapolation error: at each word, we fit a linear trajectory to the preceding hidden states of a transformer language model and measure deviation from the extrapolated path. On the Natural Stories corpus, this measure is nearly orthogonal to surprisal (r = .044) and independently predicts self-paced reading times. The effect is especially pronounced in garden-path sentences, strengthens with model scale (GPT-2 Small to Large), and replicates across architectures with different positional encoding schemes (GPT-2 vs. Pythia/RoPE). A displacement control shows the effect is not reducible to representational change magnitude: displacement and extrapolation error predict in opposite directions. These findings reveal two dissociable components of processing cost: word-level prediction error (surprisal) and sensitivity to the local momentum of the unfolding interpretation (trajectory extrapolation error).
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses negative.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05336unread
Self-supervised User Profile Generation for Personalization
Clark Mingxuan Ju, Yuwei Qiu, Tong Zhao, Neil Shah · 2026-06-05
arXiv:2606. 05336v1 Announce Type: new Abstract: Personalizing large language models (LLMs) has become a central challenge as LLMs are deployed across recommendation, search, dialogue, and content generation -- settings where the same query should yield different answers given different users.
Read next because Self-supervised User Profile Generation for Personalization overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, persona, rect, under, eval, source, rate, without. Source: arxiv cs.CL (NLP).
arXiv:2606.05336v1 Announce Type: new Abstract: Personalizing large language models (LLMs) has become a central challenge as LLMs are deployed across recommendation, search, dialogue, and content generation -- settings where the same query should yield different answers given different users. A promising route is to summarize each user's interaction history into a natural-language memory or profile and prepend it to the prompt to facilitate personalization. Existing methods learn such profile generators with explicit rewards derived from labeled downstream tasks, which are expensive and sparse as they require annotated supervision for every target task. In light of this challenge, we introduce Bidirectional User Modeling via Profiles (BUMP), a self-supervised framework that trains a profile generator without any downstream labels. Specifically, given a user's interaction history, we use GRPO to train an LLM to emit a free-form textual profile under a bidirectional in-batch ranking objective: a small LLM judge measures (i) how well the generated profile, used as a query, ranks the user's own held-out interactions above interactions from other users in the batch, and (ii) how well a held-out interaction, used as a query, ranks the user's own profile above profiles of other users. Both directions are scored with multi-positive NDCG and combined into a dense reward per rollout; other users in the batch supply free negatives, so every training example yields supervision from raw interaction logs alone. Evaluated on the LaMP benchmark, BUMP matches or outperforms closed-source APIs and prior methods relying on labeled rewards, while requiring no task label at training.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses negative, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05330unread
A Model of Multi-turn Human Persuadability Using Probabilistic Belief Tracing
Jared Moore, Noah Goodman, Nick Haber, Max Kleiman-Weiner · 2026-06-05
arXiv:2606. 05330v1 Announce Type: new Abstract: Large language models can shift human beliefs across high-stakes domains, but most persuasion studies rely on pre/post belief change.
Read next because A Model of Multi-turn Human Persuadability Using Probabilistic Belief Tracing overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, text, persona, eval, line, rate, alone, language. Source: arxiv cs.CL (NLP).
arXiv:2606.05330v1 Announce Type: new Abstract: Large language models can shift human beliefs across high-stakes domains, but most persuasion studies rely on pre/post belief change. These endpoint measures identify whether persuasion occurred, yet miss where and how beliefs moved within a dialogue. We present PERSUASIONTRACE, a framework for studying persuasion in human-LLM interaction. Built on a web-based experimental platform, PERSUASIONTRACE contributes a tool for multi-turn persuasion studies and a process-level evaluation protocol: it records multi-turn belief reports from human or simulated targets of persuasion, annotates persuader turns with rhetorical dimensions (logos/pathos/ethos), and evaluates simulators by fidelity to real human belief dynamics. Using this framework, we find that human targets group into two clusters of multi-turn belief updates and exhibit susceptibility to rhetorical strategies, and that LLMs are persuasive across generic and personalized topics, text and audio modalities, and multi-turn interactions. Prior work has chiefly used vanilla-prompted LLMs to simulate human targets, but we show that these simulators fail to replicate human belief dynamics. We introduce a Bayesian-network simulated target that maintains an explicit latent belief state over time so each persuader message yields cognitively realistic belief updates. In human-likeness evaluation, our Bayesian target scores near a human reference (81 vs 80), while baseline LLM targets score substantially lower (64). PERSUASIONTRACE reframes persuasion evaluation from endpoint movement alone to process fidelity, providing a stronger basis for scientific analysis and safer optimization of persuasive systems.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05315unread
LoRi: Low-Rank Distillation for Implicit Reasoning
Ryan Solgi, Jiayi Tian, Zheng Zhang · 2026-06-05
arXiv:2606. 05315v1 Announce Type: new Abstract: Implicit chain-of-thought (iCoT) methods aim to internalize reasoning in large language models, but often underperform explicit CoT prompting.
Read next because LoRi: Low-Rank Distillation for Implicit Reasoning overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?". Matching terms: under, eval, chain, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05315v1 Announce Type: new Abstract: Implicit chain-of-thought (iCoT) methods aim to internalize reasoning in large language models, but often underperform explicit CoT prompting. We empirically find that hidden-state reasoning trajectories exhibit low-rank structure. Motivated by this observation, we propose a low-rank distillation framework that transfers reasoning by aligning teacher and student trajectories in a shared low-rank tensor subspace using first- and second-order statistics. The resulting formulation captures the global structure of reasoning while supporting a compact latent reasoning process. We evaluate the method across multiple model families, including LLaMA and Qwen, at different scales on mathematical reasoning benchmarks. Our approach consistently improves performance, especially on challenging multi-step tasks, approaching explicit CoT accuracy and outperforming prior iCoT distillation methods.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05183unread
The Granularity Gap: A Multi-Dimensional Longitudinal Audit of Sycophancy in Gemini Models
Patrick Keough · 2026-06-05
arXiv:2606. 05183v1 Announce Type: new Abstract: Large language models are increasingly deployed as high-stakes advisors, yet standard alignment benchmarks treat sycophancy as a binary failure mode.
Read next because The Granularity Gap: A Multi-Dimensional Longitudinal Audit of Sycophancy in Gemini Models overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)", clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)". Matching terms: rect, under, alignment, correct, soft, eval, rate, control. Source: arxiv cs.CL (NLP).
arXiv:2606.05183v1 Announce Type: new Abstract: Large language models are increasingly deployed as high-stakes advisors, yet standard alignment benchmarks treat sycophancy as a binary failure mode. We introduce the Granularity Gap: coarse binary metrics mask substantial social-compliance behaviors where models capitulate to user framing, validate questionable premises, or soften factual corrections without producing overtly false outputs. We evaluate six Gemini variants across generations 2.0, 2.5, and 3.0 on 73 adversarial prompts under three guardrail conditions (Control, Simple, Protocol), yielding 8,830 graded responses. Using a 0-4 Likert scale validated against a human annotator triad (Fleiss kappa = 0.71; Cohen kappa = 0.78 vs AI consensus; 95.9 percent binary accuracy, 100 percent specificity), we quantify sycophancy as continuous rather than binary. Three findings emerge. First, 27.2 percent of responses contain substantial sycophantic content (Likert >= 2.0) and 22.7 percent reach moderate or severe levels (>= 3.0), while binary win-rate framing reports only modest failure rates; coarse metrics explain just 29 percent of graded variance. Second, generational progress is non-monotonic: Gen 2.5 regresses sharply (mean Control 2.64) relative to Gen 2.0 (1.90) and Gen 3.0 (2.01), and Gen 2.5 shows inverse scaling (Pro 1.94 worse than Flash 1.71) while Gen 3.0 restores standard scaling. Third, we document an Alignment Tax: Spearman rho = -0.63 between sycophancy and truthfulness, indicating social compliance trades against factual accuracy. Egotistical Validation prompts act as a sycophancy trap (mean 3.27), nearly double Unethical Proposals (1.72). Simple guardrails outperform elaborate Protocol scaffolding on flagship models, but distilled Gen 3.0 Flash inverts this, suggesting small models may structurally require chain-of-thought scaffolding. We release the dataset and rubric to support continuous sycophancy measurement.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses failure, adversarial, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05182unread
LANTERN: Layered Archival and Temporal Episodic Retrieval Network for Long-Context LLM Conversations
Rahul Subramani · 2026-06-05
arXiv:2606. 05182v1 Announce Type: new Abstract: Large language models discard critical details when conversation history is compacted to fit within finite context windows.
Read next because LANTERN: Layered Archival and Temporal Episodic Retrieval Network for Long-Context LLM Conversations overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, line, implement, extraction, without, full, test. Source: arxiv cs.CL (NLP).
arXiv:2606.05182v1 Announce Type: new Abstract: Large language models discard critical details when conversation history is compacted to fit within finite context windows. We present LANTERN (Layered Archival aNd Temporal Episodic Retrieval Network), a lightweight memory layer that proactively archives every conversation turn and restores relevant details after compaction via hybrid retrieval -- requiring zero LLM calls and adding fewer than 25ms of latency per turn. On 94 real multi-turn conversations (1,894 ground-truth facts, human-validated at kappa=0.81), LANTERN-Rerank recovers 78.3% of verifiable facts lost to compaction, significantly outperforming a faithful reimplementation of MemGPT's LLM-driven extraction and multi-query search pipeline (72.4%; Wilcoxon p<0.0001, 95% CI [+3.1, +8.6] pp, d=0.43) at a fraction of the inference cost. Even without the reranker, base LANTERN matches or exceeds this LLM-driven baseline (p=0.005) using zero LLM calls. When four production LLMs answer fact-bearing questions using LANTERN-restored context, accuracy improves by 8.4 percentage points on average (Wilcoxon p<0.05 for each model individually), demonstrating that the recovered context is useful across diverse model architectures. We release the full evaluation framework -- paired significance tests, failure analysis, fact-type stratification, and compaction robustness analysis -- to support reproducibility and future work.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, robustness, evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05181unread
Multi-Granularity Reasoning for Natural Language Inference
Chunling Xi, Di Liang · 2026-06-05
arXiv:2606. 05181v1 Announce Type: new Abstract: Natural Language Inference (NLI) is a fundamental task in natural language understanding that requires determining the logical relationship between a premise and a hypothesis.
Read next because Multi-Granularity Reasoning for Natural Language Inference overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, text, under, token, line, rate, trained, position. Source: arxiv cs.CL (NLP).
arXiv:2606.05181v1 Announce Type: new Abstract: Natural Language Inference (NLI) is a fundamental task in natural language understanding that requires determining the logical relationship between a premise and a hypothesis. Despite the remarkable success of transformer-based pre-trained models, most existing approaches primarily rely on the final-layer token representations, which are often insufficient for capturing the complex and hierarchical semantic interactions required for effective reasoning. In particular, fine-grained lexical cues, phrasal compositions, and higher-level contextual semantics are typically entangled or diluted in a single representation space. To address these limitations, we propose a novel \emph{Multi-Granularity Reasoning Network} (MGRN) that explicitly leverages hierarchical semantic features within an interactive reasoning space. The proposed framework mimics the human cognitive process of language understanding, which naturally progresses from shallow lexical matching to deeper semantic abstraction and logical reasoning. By integrating semantic information across multiple granularities in a progressive and structured manner, MGRN is able to uncover intricate semantic relationships underlying natural language expressions. Extensive experiments on multiple public benchmarks demonstrate that MGRN consistently outperforms strong baseline models, validating the effectiveness and robustness of the proposed approach.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, robustness, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05179unread
Efficient Punctuation Restoration via Weighted Lookahead Scoring Method for Streaming ASR Systems
Sungmook Woo, Hyungu Kang, Chanwoo Kim · 2026-06-05
arXiv:2606. 05179v1 Announce Type: new Abstract: Punctuation restoration improves ASR (Automatic Speech Recognition) readability.
Read next because Efficient Punctuation Restoration via Weighted Lookahead Scoring Method for Streaming ASR Systems overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, word, class, under, alignment, alpha, eval, token. Source: arxiv cs.CL (NLP).
arXiv:2606.05179v1 Announce Type: new Abstract: Punctuation restoration improves ASR (Automatic Speech Recognition) readability. However streaming ASR requires online decisions with limited future context. In streaming ASR, the system predicts punctuation incrementally, which makes generation-based approaches prone to latency and alignment failures under boundary-wise evaluation. This paper proposes a non-autoregressive scoring method (no free-form generation) that preserves the input transcript and makes a decision at each word boundary. Our method compares punctuation insertion hypotheses against a no-insertion baseline under a bounded K-subword-token lookahead, and calibrates decisions using a weight {\alpha} and a validation-calibrated threshold {\tau} (no parameter updates during inference). On IWSLT 2017, our scoring method achieves a 4-class macro F1 of 0.893 in the no fine-tuning setting (validation-calibrated, K=2) and 0.937 after fine-tuning (K=2), outperforming the prompt-based baseline (0.566) and a fine-tuned ELECTRA baseline (0.913) under the same lookahead budget. We analyze the impact of the lookahead budget through ablation studies on K.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, failures, evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05177unread
MCBench: A Multicontext Safety Assessment Benchmark for Omni Large Language Models
Manh Luong, Tamas Abraham, Junae Kim, Amar Kaur, Rollin Omari, Gholamreza Haffari, Trang Vu, Lizhen Qu, Dinh Phung · 2026-06-05
arXiv:2606. 05177v1 Announce Type: new Abstract: Existing multimodal safety benchmarks focus solely on visual inputs and cannot assess Omni Large Language Models (LLMs) that process vision, audio, and text.
Read next because MCBench: A Multicontext Safety Assessment Benchmark for Omni Large Language Models overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, under, eval, rate, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05177v1 Announce Type: new Abstract: Existing multimodal safety benchmarks focus solely on visual inputs and cannot assess Omni Large Language Models (LLMs) that process vision, audio, and text. We introduce MCBench, a benchmark with 1196 scenarios spanning four safety categories that require integrating multiple modalities for accurate safety assessment. Each unsafe scenario is paired with a minimally different safe counterpart to assess model sensitivity. Our evaluations of state-of-the-art models reveal significant challenges. Omni LLMs struggle with subtle or non-physical risks but perform better when salient visual or acoustic cues are present. Analysis of reasoning traces shows that, although models can extract modality-specific information, they often fail to integrate these cues effectively for safety judgments. Our findings reveal that current Omni LLMs lack robust cross-modal reasoning in safety-critical settings, underscoring the need for improved architectures and training strategies for multimodal safety.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05176unread
PEFT of SLM for Telecommunications Customer Support: A Comparative Study of LoRA Configurations with Energy Consumption Analysis
Lucas Tamic, Ilan Jaffeux-Cheniout, Xavier Marjou · 2026-06-05
arXiv:2606. 05176v1 Announce Type: new Abstract: While large language models (LLMs) show strong performance in natural language understanding and generation, their evaluation and adaptation to domain-specific constraints in telecommunications customer support remain limited.
Read next because PEFT of SLM for Telecommunications Customer Support: A Comparative Study of LoRA Configurations with Energy Consumption Analysis overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, under, eval, assistant, line, alone, lora, qwen2. Source: arxiv cs.CL (NLP).
arXiv:2606.05176v1 Announce Type: new Abstract: While large language models (LLMs) show strong performance in natural language understanding and generation, their evaluation and adaptation to domain-specific constraints in telecommunications customer support remain limited. In addition, data sovereignty, regulatory constraints, and the handling of sensitive customer and network information complicate the use of externally hosted foundation models in this domain. We present a systematic study of parameter-efficient fine-tuning (PEFT) using Low-Rank Adaptation (LoRA) applied to Qwen2.5-3B to build a domain-specific conversational assistant. We introduce a combinatorial synthetic data generation approach based on a glossary of 52 industry-specific terms, producing approximately 30,000 training examples across 1,560 distinct problem scenarios via a generative pipeline powered by Gemini 2.0 Flash. We evaluate 16 LoRA configurations by varying hyperparameters and target modules. Our evaluation extends beyond standard metrics by incorporating energy consumption analysis and qualitative assessment using an LLM-as-a-judge framework with GPT-5.2 and Claude 4.5 Sonnet. Results show a clear divergence between quantitative and qualitative performance: models achieving the lowest validation loss do not necessarily obtain the best human-aligned rankings. The best validation loss (0.5024) ranks only 6th-7th in qualitative evaluation, while the worst loss (0.6807) ranks first according to both judges. This work contributes (1) a combinatorial method for synthetic dataset construction, (2) insights into the impact of target module selection for LoRA injection, (3) evidence that validation loss alone is insufficient for selecting fine-tuning configurations in conversational AI, and (4) an energy-performance trade-off analysis for sustainable LLM deployment.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05175unread
Generic Triple-Latent Compression with Gated Associative Retrieval
Liu Xiao · 2026-06-05
arXiv:2606. 05175v1 Announce Type: new Abstract: We study generic triple-latent sequence models that maintain a running token state and compressed pair-memory pathway to capture higher-order token interactions without benchmark-specific parsing.
Read next because Generic Triple-Latent Compression with Gated Associative Retrieval overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: text, eval, token, line, implement, without, language, model. Source: arxiv cs.CL (NLP).
arXiv:2606.05175v1 Announce Type: new Abstract: We study generic triple-latent sequence models that maintain a running token state and compressed pair-memory pathway to capture higher-order token interactions without benchmark-specific parsing. The triple-latent family improves a small Transformer baseline on byte-level WikiText-2 and on a tokenizer-based MiniMind language-model benchmark, while a recall-focused gated key-value retrieval extension improves associative recall but remains seed-sensitive and much slower in the current reference implementation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 100arxiv cs.CL (NLP)arxiv:2606.05173unread
Predict and Reconstruct: Joint Objectives for Self-Supervised Language Representation Learning
Aimen Boukhari · 2026-06-05
arXiv:2606. 05173v1 Announce Type: new Abstract: Masked language modelling (MLM) has been the dominant pre-training objective for text encoders since BERT, yet it encourages representations that are strongly anchored to surface-form token identity rather than deeper semantic structure.
Read next because Predict and Reconstruct: Joint Objectives for Self-Supervised Language Representation Learning overlaps with clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, strong, text, under, token, line, rate, alone. Source: arxiv cs.CL (NLP).
arXiv:2606.05173v1 Announce Type: new Abstract: Masked language modelling (MLM) has been the dominant pre-training objective for text encoders since BERT, yet it encourages representations that are strongly anchored to surface-form token identity rather than deeper semantic structure. Inspired by the success of Joint Embedding Predictive Architectures (JEPA) (LeCun, 2022) in vision and audio, we propose a hybrid pre-training objective that combines a JEPA-style latent-space prediction loss with a standard MLM objective over a single shared encoder. A learnable scalar parameter continuously balances the two objectives during training. We pre-train both a hybrid model and a pure-MLM baseline on English Wikipedia using identical architectures and compute budgets (NVIDIA H100). Extensive representation analysis across five GLUE benchmarks (SST-2, MRPC, MNLI, CoLA, STS-B) using four pooling strategies reveals that the hybrid encoder produces significantly more uniform embeddings (uniformity less than -0.16 vs -0.05 for MLM), exhibits richer spectral geometry under max pooling, encodes less surface-level lexical information, and achieves a better semantic-to-lexical balance. Despite similar linear-probe downstream accuracy, the geometric differences are consistent and significant, suggesting that the JEPA predictive objective reshapes the latent space in ways that standard accuracy metrics alone cannot capture.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits at 23.5% when a co-trained partner learns <A>..., vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
source ↗open
score 96arxiv cs.LG (Machine Learning)arxiv:2606.04075unread
Large Language Models Hack Rewards, and Society
Wei Liu, Xinyi Mou, Hanqi Yan, Zhongyu Wei, Yulan He · 2026-06-05
arXiv:2606. 04075v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a dominant post-training paradigm, enabling large language models (LLMs) to learn from rewards.
Read next because Large Language Models Hack Rewards, and Society overlaps with clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation", experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: rate, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04075v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a dominant post-training paradigm, enabling large language models (LLMs) to learn from rewards. We observe that societal regulations are structurally similar to reward functions. They define measurable outcomes, thresholds, and exceptions, while often leaving institutional intent only partially specified. We hypothesise that the RL training process may exploit these gaps and therefore ask whether models' well-known tendency to hack reward functions during RL can scale into a more consequential failure mode named societal hacking: discovering loopholes in the rules society runs on. To study this phenomenon, we introduce SocioHack, a sandbox of 72 societal environments, and find that within these environments, reward hacking naturally emerges and leads to regulatory loophole discovery. Models learn to hack the social rules and generate strategies that remain technically compliant while defeating regulatory intent, and current LLM safeguards provide only limited mitigation. Therefore, collecting in-the-wild feedback for model training requires greater caution, and we need a next-generation post-training paradigm for safely iterating LLMs in real society.=
Potential threat/caveat for clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)": this item discusses failure.
source ↗open
score 96arxiv cs.LG (Machine Learning)arxiv:2606.04031unread
Pseudospectral Bounds for Transient Amplification in Coupled Gradient Descent
Ahanaf Hasan Ariq · 2026-06-05
arXiv:2606. 04031v1 Announce Type: new Abstract: Coupled gradient descent--where the update of one parameter block depends on another--underlies bilevel optimization, two-time-scale stochastic approximation, and adversarial training.
Read next because Pseudospectral Bounds for Transient Amplification in Coupled Gradient Descent overlaps with clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Can capability be taught through another persona?". Matching terms: under, line, another. Source: arxiv cs.LG (Machine Learning).
arXiv:2606.04031v1 Announce Type: new Abstract: Coupled gradient descent--where the update of one parameter block depends on another--underlies bilevel optimization, two-time-scale stochastic approximation, and adversarial training. When the coupled Jacobian is block-triangular, asymptotic stability is governed by the spectral radii of the diagonal blocks, yet transient amplification before convergence can be arbitrarily large due to non-normality. We develop a sharp pseudospectral theory for such block-triangular Jacobians, proving that the Kreiss constant satisfies $K(J) \leq 2/(1-\gamma) + \|C\|/(4(1-\gamma))$ when the diagonal blocks are symmetric with spectral radii at most $\gamma < 1$, and we establish matching minimax lower bounds. We characterize the critical coupling threshold for spectral instability and extend the analysis to nearly self-referential systems via a Neumann-series perturbation framework. As a consequence, we obtain a finite-horizon iteration-complexity bound of $O(K(J)^2 \log(1/\delta))$ for stochastic coupled descent. Framed as scaling laws for non-stationary two-time-scale optimization, our results expose a non-asymptotic, instance-dependent regime of high-dimensional learning dynamics that is invisible to spectral-radius analysis. Experiments on linear-quadratic problems, IQC-based comparisons, and neural-network training confirm the theory.
Potential threat/caveat for clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)": this item discusses adversarial.
source ↗open
score 96arxiv cs.CR (Cryptography and Security)arxiv:2606.05418unread
A formal framework for the economic security of DeFi compositions
Massimo Bartoletti, Riccado Marchesin, Roberto Zunino · 2026-06-05
arXiv:2606. 05418v1 Announce Type: new Abstract: Decentralized Finance (DeFi) services are usually constructed by composing a variety of smart contracts.
Read next because A formal framework for the economic security of DeFi compositions overlaps with experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?", experiment "#351 follow-up: broader-vocab position-0 sweep at T=1.0 + position-1 suffix isolation", experiment "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: chain, position, model. Source: arxiv cs.CR (Cryptography and Security).
arXiv:2606.05418v1 Announce Type: new Abstract: Decentralized Finance (DeFi) services are usually constructed by composing a variety of smart contracts. While composability is a key driver of the success of DeFi, it also creates security risks: adversaries may exploit interactions between newly deployed contracts and the pre-existing ones to inflict economic losses. We introduce MEV non-interference, a formal security notion for DeFi composability requiring that the maximal extractable value from a set of newly deployed contracts is not increased by interactions with the existing blockchain state. To support this notion, we define local MEV, a novel measure of economic attacks that focusses on the loss of a given set of victim contracts. We study two adversarial models, with bounded and unbounded wealth, and establish sufficient conditions and locality principles that enable modular reasoning about secure composability. We apply the framework to representative DeFi compositions, including exchanges, AMMs, options, lending pools, routers, and arbitrage contracts, showing how it distinguishes secure compositions from vulnerable ones. Our results provide a formal foundation for reasoning about the economic security of DeFi compositions.
Potential threat/caveat for experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?": this item discusses adversarial.
source ↗open
score 80arxiv stat.ML (Machine Learning)arxiv:2606.05242unread
Deterministic Envelopes for Tamed SGLD: Decoupling Stochastic-Gradient Noise and Localizing Taming
Yiwei Zhou, Ziheng Chen · 2026-06-05
arXiv:2606. 05242v1 Announce Type: new Abstract: Stochastic-gradient Langevin algorithms often use tamed denominators to stabilize non-globally Lipschitz drifts.
Read next because Deterministic Envelopes for Tamed SGLD: Decoupling Stochastic-Gradient Noise and Localizing Taming overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone". Matching terms: soft, control. Source: arxiv stat.ML (Machine Learning).
arXiv:2606.05242v1 Announce Type: new Abstract: Stochastic-gradient Langevin algorithms often use tamed denominators to stabilize non-globally Lipschitz drifts. This paper shows that when the denominator depends on the same stochastic-gradient realization as the numerator, the taming step changes the stochastic oracle itself and can create a stationary bias even if the original stochastic gradient is unbiased. We propose a structure-preserving framework for designing tamed denominators. It fixes the denominator before the oracle noise is sampled and uses localized deterministic envelopes to avoid unnecessary taming in typical regions. These kernels keep the stabilizing effect of taming while avoiding the bias introduced by a gradient-dependent denominator. Our theory explains how the stationary error splits into the bias caused by oracle-dependent taming and the remaining error introduced by deterministic stabilization. Within this deterministic-envelope family, the analysis identifies a far-tail condition that explains the limitation of local soft envelopes and motivates a hybrid member: soft in the typical region, but protected by hard-tail control on rare excursions. Experiments confirm the predicted stationary distortions of random denominators, the bias reduction of deterministic-envelope designs, and the stabilizing effect of the hybrid construction.
Potential threat/caveat for clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)": this item discusses limitation, bias.
source ↗open

Methods

score 38M7 QA inline RSS threat sourceunread
Artifact verification caveats for Sagan clean results
M7 QA · No release date
This paper studies failure modes and caveats when Sagan creates a clean result only after verifying an artifact row. It proposes benchmark checks for artifact verification, clean-result review comments, and negative controls.
My work produces clean results across multiple experiments (marker leakage, language spill, backdoor triggers, alignment collapse), so understanding the verification failure modes described here could affect how much weight I place on MODERATE- or LOW-confidence findings already in my context.
This paper studies failure modes and caveats when Sagan creates a clean result only after verifying an artifact row. It proposes benchmark checks for artifact verification, clean-result review comments, and negative controls.
Potential threat/caveat for experiment "Add C2 control arm (donor sees marker_B without marker_A) to disambiguate paired-marker binding from marker_B leaking alone": this item discusses failure, caveat, caveats, negative, benchmark.
source ↗open