Active sources: 7. Sources represented in this queue: 3. The cron runs daily at 06:00 server time; arxiv RSS is often empty on weekends.
Linked to your results
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22093unread
Knowledge Graph Re-engineering Along the Ontological Continuum (extended version)
Enrico Daga, Valentina Tamma, Terry Payne · 2026-05-23
arXiv:2605. 22093v1 Announce Type: new Abstract: Knowledge graphs have become the primary vehicle for data integration and are critical to the success of modern AI, but the diversity of KG modelling practices, from lightweight vocabularies to richly axiomatised ontologies, makes integration and reuse expensive and brittle.
Read next because Knowledge Graph Re-engineering Along the Ontological Continuum (extended version) 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)", experiment "Implement Chen et al. persona-vector extraction recipe and compare to project's centroid-difference recipe", 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, compare, without, full, capability, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22093v1 Announce Type: new Abstract: Knowledge graphs have become the primary vehicle for data integration and are critical to the success of modern AI, but the diversity of KG modelling practices, from lightweight vocabularies to richly axiomatised ontologies, makes integration and reuse expensive and brittle. This challenge is particularly acute in neuro-symbolic AI, where bridging neural and symbolic components depends on the ability to reengineer KGs to fit new requirements; GenAI now offers unprecedented automation capability, but without a principled understanding of the KG space, such automation remains conceptually ungrounded. We introduce the ontological continuum as that missing conceptualisation, a theoretical construct a theoretical construct whose characterisation framework is defined by two orthogonal distinctions: semantics vs pragmatics, and properties vs affordances; together these define a vocabulary to describe, compare, navigate, and transform KGs across the full range of modelling practices. The methodological stance is empirical: rather than prescribing how KGs should be modelled, the continuum aims to define a theory of the existent, derived from observation of real-world KG engineering practices and whose structure can be made formally explicit, for example, through Formal Concept Analysis (FCA). We ground the vision through a case study on provenance knowledge, showing how a single concern manifests differently across the continuum. We articulate five open research challenges and invite the community to develop the ontological continuum as a shared research agenda.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21997unread
The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems
Yohei Nakajima · 2026-05-23
arXiv:2605. 21997v1 Announce Type: new Abstract: Most agent frameworks are built around the language model: a conversation loop comes first, then tools, then rules, and finally a logging layer bolted on for observability, with state persisted as retrievable "memory.
Read next because The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems 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, prefix, source, line, rate, project, without. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21997v1 Announce Type: new Abstract: Most agent frameworks are built around the language model: a conversation loop comes first, then tools, then rules, and finally a logging layer bolted on for observability, with state persisted as retrievable "memory." We describe ActiveGraph, a runtime that inverts this arrangement. The append-only event log is the source of truth; the working graph is a deterministic projection of that log; and behaviors--ordinary functions, classes, LLM-backed routines, or logic attached to typed edges--react to changes in the graph and emit new events. No component instructs another; coordination happens entirely through the shared graph. This single design decision yields three properties that retrieval-and-summarization memory systems do not provide: deterministic replay of any run from its log, cheap forking that branches a run at any event without re-executing the shared prefix, and end-to-end lineage from a high-level goal down to the individual model call that produced each artifact. We present the architecture, a determinism contract that makes replay sound, and a worked diligence example whose full causal structure is reconstructable from the log alone. We discuss--without claiming to demonstrate--why this substrate is unusually well suited to self-improving agents, and how it extends the BabyAGI lineage and prior graph-memory research.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21832unread
FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation
Xinhang Yuan, Zexi Huang, Anjia Cao, Xudong Lu, Zikai Wang, Penghao Zhou, Chang Liu, Wentao Guo, Qinglei Wang · 2026-05-23
arXiv:2605. 21832v1 Announce Type: new Abstract: Modern recommender systems rely heavily on ID-based collaborative filtering: each item is represented by a unique ID embedding that accumulates collaborative signals from user interactions.
Read next because FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, token, line, full, trained, stage, candidate. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21832v1 Announce Type: new Abstract: Modern recommender systems rely heavily on ID-based collaborative filtering: each item is represented by a unique ID embedding that accumulates collaborative signals from user interactions. Livestreaming recommendation, however, faces a unique challenge in this paradigm: a live room typically broadcasts for only tens of minutes, so its item ID remains poorly learned in a persistent cold-start state and ID-centric ranking models fail to generalize. We present FLUID, the first framework to fully retire the candidate-side item ID from a production-scale livestreaming ranker. FLUID couples a cross-domain multimodal encoder, jointly trained on short videos and livestreams to produce discrete hierarchical codes (LUCID), with a late-fusion, ID-free design that injects slice-level and room-level LUCID as independent tokens, stabilized by a staged warmup under online incremental training. Deployed on our industrial livestreaming recommenders with a cross-platform combined user base of over one billion globally, FLUID delivers significant online gains of +0.55% Quality Watch Duration, +2.05% Cold-Start Room Views, and +0.05% Active Hours.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21825unread
Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks
Haichao Miao, Zhimin Li, Kuangshi Ai, Kaiyuan Tang, Chaoli Wang, Peer-Timo Bremer, Shusen Liu · 2026-05-23
arXiv:2605. 21825v1 Announce Type: new Abstract: The ability to inspect, interpret, and communicate complex data is crucial for virtually any scientific endeavor, but often requires significant expertise outside the core domain ranging from data management and analysis to visualization design and implementation.
Read next because Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, implement, stage, test, lora, completion. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21825v1 Announce Type: new Abstract: The ability to inspect, interpret, and communicate complex data is crucial for virtually any scientific endeavor, but often requires significant expertise outside the core domain ranging from data management and analysis to visualization design and implementation. We present an end-to-end agentic harness that, based on only the data and a high level description of the tasks, independently designs custom visual analysis applications (VIS apps). This represents an important step towards a general AI co-scientist envisioned by many as an autonomous system that can autonomously execute long horizon tasks based on high-level directions. Our proposed VIS co-scientist is an essential component of this broader AI co-scientist vision: a harness that can autonomously analyze data and design visualization solutions using a collection of agents and specialized skills that coordinate exploratory analysis, plan, configure the environment, implement, validate the interface, and most importantly evaluate the overall task completion. Each stage produces document and instruction artifacts that guide downstream work and enable iterative refinement. We validate this approach on IEEE SciVis Contests spanning multiple science and engineering fields. These contests serve as ideal proving grounds because they encode real-world complexity: ambiguous requirements, diverse data modalities, design trade-offs, and task-driven validation. Given only the data and target tasks, our system autonomously produces functional single-page VIS Apps with verified linked-view behavior, highly customized to domain experts' specified tasks and needs.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21695unread
The Impact of AI Usage and Informativeness on Skill Development in Logical Reasoning
Shang Wu, Hongyu Yao, Catarina Belem, Shuyuan Fu, Mark Steyvers, Padhraic Smyth · 2026-05-23
arXiv:2605. 21695v1 Announce Type: new Abstract: Artificial intelligence (AI) is being increasingly integrated into human problem-solving, yet its effects on individual skill development remain unclear.
Read next because The Impact of AI Usage and Informativeness on Skill Development in Logical Reasoning overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, under, rate, control, without. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21695v1 Announce Type: new Abstract: Artificial intelligence (AI) is being increasingly integrated into human problem-solving, yet its effects on individual skill development remain unclear. We examine how both AI usage and informativeness can shape learning in the context of a controlled logical reasoning task with on-demand access to AI assistance. We find that greater AI usage is associated with weaker skill development: heavy AI users underperform relative to comparable peers, whereas light AI users perform similarly to matched users who do not use AI. We also find in our study that these patterns are mediated by AI informativeness. Low-information AI neither improves immediate performance nor preserves performance after AI assistance is removed, and is linked to weaker learning overall. On the other hand, high-information AI was found to improve short-run performance without reducing post-AI outcomes on average in our experiments, but with heterogeneous effects. Our findings in general suggest that AI can, depending on context, either complement human skill development by amplifying independent reasoning or can act as a substitute that undermines such reasoning, with the implication that regulating AI access and usage will be important for promoting skill development in the presence of AI assistance.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21683unread
Investigating Concept Alignment Using Implausible Category Members
Sunayana Rane, Brenden M. Lake, Thomas L. Griffiths · 2026-05-23
arXiv:2605. 21683v1 Announce Type: new Abstract: Developing AI systems with a human-like understanding of everyday concepts is a key step towards developing safe, reliable systems whose behavior makes sense to humans.
Read next because Investigating Concept Alignment Using Implausible Category Members 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: word, class, under, alignment, rate, compare, full, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21683v1 Announce Type: new Abstract: Developing AI systems with a human-like understanding of everyday concepts is a key step towards developing safe, reliable systems whose behavior makes sense to humans. When probing concept understanding, asking questions about plausible category members (e.g., "Is a car a vehicle?") is likely to recall patterns in the model's vast training data. We pursue an alternative strategy, characterizing the boundaries of conceptual categories by asking about implausible category members (e.g., "Is an olive a vehicle?") to probe the kind of concept-level knowledge we take for granted in fellow humans. We characterize concept boundaries for a set of fundamental concepts by studying AI systems' assignments of objects to superordinate categories from a classic psychological study by Rosch and Mervis, as well as their assignments of the same objects to mismatched superordinate categories. We compare these assignments to those made by human participants on the full range of within-category and cross-category assignment tasks. Our results reveal a range of concepts for which which models differ in meaningful and surprising ways from humans, including treating "words" as belonging to categories like "vehicles" and "clothing," identifying several "vegetable" category members as "fruit," and assigning exemplars from non-weapon categories to the "weapons" category. We also demonstrate how these instances of concept misalignment translate into problematic downstream behavior with implications for AI safety.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21648unread
Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos
Lucas Fernandez Sarmiento · 2026-05-23
arXiv:2605. 21648v1 Announce Type: new Abstract: We develop a mean-field theory of dropout as a perturbation of critical signal propagation at the edge of chaos.
Read next because Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos 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, alignment, rate, test. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21648v1 Announce Type: new Abstract: We develop a mean-field theory of dropout as a perturbation of critical signal propagation at the edge of chaos. Dropout shifts the perfect-alignment fixed point, making the depth scale for information propagation finite even at critical initialization. We derive critical and crossover scaling laws for correlation decay and establish that smooth activations and kinked, ReLU-like activations constitute distinct universality classes, with different critical exponents and a universal two-parameter scaling collapse in detuning and dropout strength. The distinction traces to the analytic structure of the correlation map: smooth activations admit a Taylor expansion near perfect alignment, while kinked activations develop a branch point with universal non-analyticity. As a corollary, the framework yields saturated dropout profiles under fixed budget; a rank-flow tie-breaker then selects front-loaded schedules, substantially reducing held-out test loss at no extra computational cost, with accuracy gains as a consistent secondary effect. We test the predictions in MLPs and Vision Transformers and discuss CNN/ResNet extensions.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21617unread
$\textit{BlockFormer}$ : Transformer-based inference from interaction maps
Elo\"ise Touron, Pedro L. C. Rodrigues, Julyan Arbel, Nelle Varoquaux, Michael Arbel · 2026-05-23
arXiv:2605. 21617v1 Announce Type: new Abstract: Inference from interaction maps, such as centromere identification from genome-wide chromosome conformation capture techniques -- notably Hi-C -- can be formulated as a generic inverse problem: infer a set of parameters given a map summarizing pairwise interactions between entities through blocks of variable numbers and sizes.
Read next because $\textit{BlockFormer}$ : Transformer-based inference from interaction maps overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, alignment, rate, position. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21617v1 Announce Type: new Abstract: Inference from interaction maps, such as centromere identification from genome-wide chromosome conformation capture techniques -- notably Hi-C -- can be formulated as a generic inverse problem: infer a set of parameters given a map summarizing pairwise interactions between entities through blocks of variable numbers and sizes. In this work, we introduce a data-driven approach that leverages shared structure between these maps, such as global alignment between localized patterns, while handling the variability in number and size of entities arising in real-world data. Our approach relies on a transformer architecture capable of handling such variability and a custom simulator to generate abundant, yet computationally cheap synthetic data for training. Applied to the problem of centromere localization, the method accurately recovers their genomic positions across a wide range of species of various genome sizes.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21558unread
From Parameters to Data: A Task-Parameter-Guided Fine-Tuning Pipeline for Efficient LLM Alignment
Hao Chen, Qi Zhang, Liyao Li, Zhanming Shen, Wentao Ye, Lirong Gao, Ningtao Wang, Xing Fu, Xiaoyu Shen, Junbo Zhao · 2026-05-23
arXiv:2605. 21558v1 Announce Type: new Abstract: Adapting Large Language Models (LLMs) to specialized domains typically incurs high data and computational overhead.
Read next because From Parameters to Data: A Task-Parameter-Guided Fine-Tuning Pipeline for Efficient LLM Alignment overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, alignment, line, rate, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21558v1 Announce Type: new Abstract: Adapting Large Language Models (LLMs) to specialized domains typically incurs high data and computational overhead. While prior efficiency efforts have largely treated data selection and parameter-efficient fine-tuning as isolated processes, our empirical analysis suggests they may be intrinsically coupled. We posit the Strong Map Hypothesis: a sparse subset of attention heads plays a dominant role in task-specific adaptation, acting as keys that unlock specific data patterns. Building on this observation, we propose From Parameters to Data (P2D), a unified framework that leverages these task-sensitive attention heads as a dual compass for both sample mining and structural pruning. To rigorously quantify the total pipeline cost, we introduce the Alignment Efficiency Ratio (AER) metric for both selection latency and training time. Mechanistically, P2D identifies critical heads via a lightweight proxy and uses them as a functional filter to curate high-affinity data, establishing a synergistic pipeline. Empirically, by updating merely 10% of attention heads on 10% of the data, P2D achieves an 8.3 pp performance gain over strong baselines and delivers a 7.0x end-to-end time speedup. These results validate that precise parameter-data synchronization eliminates redundancy, offering a new paradigm for efficient alignment.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21553unread
TONIC: Token-Centric Semantic Communication for Task-Oriented Wireless Systems
Sige Liu, Kezhi Wang · 2026-05-23
arXiv:2605. 21553v1 Announce Type: new Abstract: Tokens are becoming the basic units through which foundation models represent and process information for understanding and inference.
Read next because TONIC: Token-Centric Semantic Communication for Task-Oriented Wireless Systems 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, source, token, line, full, completion. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21553v1 Announce Type: new Abstract: Tokens are becoming the basic units through which foundation models represent and process information for understanding and inference. However, traditional wireless communication, centered on bit-level fidelity, faces a mismatch between what is transmitted reliably and what downstream models actually consume. This mismatch calls for a communication design that directly accounts for token-level task relevance and downstream model requirements, rather than treating all transmitted bits as equally important. In this paper, we propose TONIC, a token-centric semantic communication framework for task-oriented wireless systems. The transmitter converts each source sample into a sequence of tokens, estimates token-level task relevance, and allocates protection through utility-aware unequal error protection under a fixed channel-use budget. At the receiver, token-level confidence is used to gate unreliable decisions, turning harmful substitutions into recoverable erasures before a Transformer-based completion model restores the masked tokens for final task inference. Our framework combines transmitter-side semantic-aware protection with receiver-side confidence-aware gating in a modular and interpretable architecture, rather than relying solely on fully black-box end-to-end learning. We further establish a utility-aware Bayes-risk interpretation for the receiver-side gating rule and study its interaction with unequal protection and completion. Experimental results on image classification show that TONIC consistently outperforms separation-based schemes, the pixel-domain DeepJSCC baseline, and token-domain baselines under matched communication budgets over AWGN, Rayleigh, and Rician channels.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21539unread
DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Models
Xuyang Zhong, Qizhang Li, Yiwen Guo, Chen Liu · 2026-05-23
arXiv:2605. 21539v1 Announce Type: new Abstract: We propose DualOptim+, a novel optimization framework for improving machine unlearning in large language models.
Read next because DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language Models overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, alignment, rate, without, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21539v1 Announce Type: new Abstract: We propose DualOptim+, a novel optimization framework for improving machine unlearning in large language models. It introduces a base state to capture common representations shared by forgetting and retaining objectives and delta states to preserve objective-specific residuals. This architecture allows the optimizer to adaptively bridge shared and decoupled states based on the directional conflict between forgetting and retaining gradients. We further introduce DualOptim+ 8bit, a quantized variant that reduces memory overhead without compromising performance. Extensive experiments across fictitious and real-world unlearning, safety alignment, and multi-task learning tasks demonstrate that DualOptim+ consistently achieves a superior trade-off between different objectives. Codes are available at https://github.com/CityU-MLO/DualOptimPlus.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21515unread
Predicting Performance of Symbolic and Prompt Programs with Examples
Chengqi Zheng, Keya Hu, Shuzhi Liu, Tao Wu, Kevin Ellis, Yewen Pu · 2026-05-23
arXiv:2605. 21515v1 Announce Type: new Abstract: LLM prompting is widely used for naturally stated tasks, yet it is unreliable it may succeed on a few test cases but fail at deployment time.
Read next because Predicting Performance of Symbolic and Prompt Programs with Examples 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, correct, test, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21515v1 Announce Type: new Abstract: LLM prompting is widely used for naturally stated tasks, yet it is unreliable it may succeed on a few test cases but fail at deployment time. We study performance prediction: given a program, either symbolic (e.g. Python) or a prompt executed on an LLM, and a few in-domain examples, predict its performance on unseen tasks from the same domain. We use a simple coin-flip model, treating each pass/fail program execution as a Bernoulli random variable, whose success probability is the programs unknown performance. In this model, performance depends entirely on: 1) the observed execution outcomes on test cases, and 2) a prior over performances. We compile empirical performance priors from a corpus of diverse programs and tasks, and find that performance for symbolic programs (e.g., Python) are all or nothing, while prompt programs have a diffuse prior with many nearly-correct programs. This difference explains why a few passing tests can certify symbolic programs but not prompt programs. Building on this insight, we develop RAP (Retrieved Approximate Prior), which retrieves similar tasks and prompt programs from an existing corpus to construct a proxy prior, which is then used to predict performance. We show RAP achieves solid performances.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21494unread
Double descent for least-squares interpolation on contaminated data: A simulation study
Tino Werner · 2026-05-23
arXiv:2605. 21494v1 Announce Type: new Abstract: Overparametrized models can exhibit an excellent generalization performance, although they should be prone to overfitting according to classical statistical theory.
Read next because Double descent for least-squares interpolation on contaminated data: A simulation study 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, good, line, compare, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21494v1 Announce Type: new Abstract: Overparametrized models can exhibit an excellent generalization performance, although they should be prone to overfitting according to classical statistical theory. The discovery of the "double descent", indicating that the generalization error decreases after a certain model complexity has been reached, opened a new line of research. Robust statistics considers statistical estimation on contaminated data, which, due to assumptions that do not hold on real data, let data points appear as outliers w.r.t. the assumed "ideal" distribution, potentially severely distorting any classical estimator. We address the question whether a double descent phenomenon can be observed in a linear regression setting with contaminated training data. We compare the performance of the highly non-robust least-squares interpolation estimator with several robust alternatives. It turns out that large overparametrization indeed allows for a double descent phenomenon, resulting in a very good generalization performance of the least-squares interpolator, surpassing that of the robust alternatives.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21490unread
Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding
Danny Butvinik (NICE Actimize), Yonit Marcus (NICE Actimize), Nitzan Tal (NICE Actimize), Gabrielle Azoulay (NICE Actimize) · 2026-05-23
arXiv:2605. 21490v1 Announce Type: new Abstract: We introduce the Temporal Contrastive Transformer (TCT), a representation learning framework designed to capture contextual temporal dynamics in sequences of financial transactions.
Read next because Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, class, rect, eval, line, rate. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21490v1 Announce Type: new Abstract: We introduce the Temporal Contrastive Transformer (TCT), a representation learning framework designed to capture contextual temporal dynamics in sequences of financial transactions. The model is trained using a self-supervised contrastive objective to produce embeddings that encode behavioral patterns over time, with the goal of supporting downstream fraud detection tasks. We evaluate TCT in a realistic setting by using the learned embeddings as input features to a gradient boosting classifier. Experimental results show that embeddings alone achieve meaningful predictive performance (AUC 0.8644), indicating that the model captures non-trivial temporal structure. However, when combined with domain-engineered features, no measurable improvement is observed over the baseline (AUC 0.9205 vs. 0.9245), suggesting that the learned representations largely overlap with existing feature abstractions. These findings position TCT as a promising representation learning approach that captures relevant behavioral signal, while highlighting the challenges of achieving additive value over strong domain features. The results reflect an intermediate stage in the development of temporal representation learning for financial crime detection and motivate further research on model architecture, training objectives, and integration strategies. At this early stage, achieving performance comparable to a strong feature-engineered baseline is itself a meaningful outcome, indicating that learned representations approximate domain-specific features without manual engineering. While not yet production-ready, these results point to a promising direction for reducing reliance on feature engineering in financial crime detection.
- score 94arxiv cs.AI (Artificial Intelligence)arxiv:2605.21623unread
The Shape of Testimony: A Scalable Framework for Oral History Archive Comparison
Itamar Trainin, Renana Keydar, Amit Pinchevski · 2026-05-23
arXiv:2605. 21623v1 Announce Type: new Abstract: Researchers in Holocaust studies have often distinguished between two styles of oral survivor testimony: the USC Shoah Foundation's interviews tend to follow a structured, interviewer-guided format, whereas the Yale Fortunoff Video Archive generally favors a more free-form, open-ended style.
Read next because The Shape of Testimony: A Scalable Framework for Oral History Archive Comparison 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 "Test FR↔IT bystander-spill symmetry at multi-seed + 5 phrasings — pooled-rate vs per-phrasing asymmetry from #239 fact-check". Matching terms: rate, test, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21623v1 Announce Type: new Abstract: Researchers in Holocaust studies have often distinguished between two styles of oral survivor testimony: the USC Shoah Foundation's interviews tend to follow a structured, interviewer-guided format, whereas the Yale Fortunoff Video Archive generally favors a more free-form, open-ended style. This distinction has influenced both scholarly research and the development of later archives. In this study, we critically examine that claim by conducting a large-scale computational analysis of more than 1,600 testimonies from both collections. Leveraging discourse segmentation, topic modeling, and large language model (LLM) based analysis, we quantify the "structuredness" level of testimonies through topic coherence, interviewer-survivor dynamics, and the distribution of question types. Our results generally corroborate the structural differences identified in earlier research, while also revealing significant overlaps between the collections, both within individual interviews and across common narrative patterns. This complicates the simple "structured vs. free-form" dichotomy often applied to these oral histories. Beyond revisiting a foundational claim in Holocaust studies, our work provides a scalable, replicable framework for comparative corpus analysis. As a proof of concept, it suggests broader applications for digital oral history, narrative analysis, and the design of citizen-science annotation platforms.
- score 46arxiv cs.LG (Machine Learning)arxiv:2605.21568unread
Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Model
Jack Kendall · 2026-05-23
arXiv:2605. 21568v1 Announce Type: new Abstract: In this work, we extend the Equilibrium Propagation framework to skew-gradient systems and show an equivalence between deep Energy-Based Models and Hamiltonian neural networks.
Read next because Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Model overlaps with 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: model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21568v1 Announce Type: new Abstract: In this work, we extend the Equilibrium Propagation framework to skew-gradient systems and show an equivalence between deep Energy-Based Models and Hamiltonian neural networks. We focus on networks of diffusively coupled Fitzhugh-Nagumo neurons as a prototypical example. We show that since stationary solutions of the Fitzhugh-Nagumo model are described by self-adjoint operators, the methods of equilibrium propagation for performing credit assignment can be applied. Furthermore, for Fitzhugh-Nagumo networks with the topology of a deep residual network, we show that the steady state solutions admit a (spatial) Hamiltonian, and thus the methods of Hamiltonian Echo Backpropagation can be applied. We end by deriving an explicit layer-wise Hamiltonian recurrence relation governing inference for stationary solutions of both deep Fitzhugh-Nagumo networks and deep Energy-Based Models.
Threats and caveats
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22106unread
ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning
Yeqiu Chen, Ziyan Liu, Zhenxin Huang, Runquan Gui, Hong Wang, Lei Liu · 2026-05-23
arXiv:2605. 22106v1 Announce Type: new Abstract: Recent progress in LLM reasoning has increasingly shifted from single-pass generation to explicit search over intermediate reasoning states.
Read next because ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM 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 "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)", 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, width, token, full. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22106v1 Announce Type: new Abstract: Recent progress in LLM reasoning has increasingly shifted from single-pass generation to explicit search over intermediate reasoning states. Tree-of-Thoughts (ToT) organizes inference to tree-structured search with branching and backtracking, but it substantially amplifies the Key--Value (KV) cache: retaining KV states for a frontier of partial trajectories quickly becomes a memory bottleneck that limits throughput and constrains search depth and width under fixed hardware budgets. We address this challenge by observing that KV reuse in ToT-style inference is governed by search dynamics: near-term decoding depends primarily on the active branch and its ancestors, whereas inactive subtrees have low short-term reuse probability yet must remain recoverable for backtracking. Motivated by this, we propose ArborKV, a structure-aware eviction framework that couples a lightweight value estimator with a tree-aware allocation policy, and performs purely token-extractive eviction with lazy rehydration to support revisits. Experiments on ToT-style reasoning benchmarks show that ArborKV achieves up to ~4x peak KV-memory reduction while preserving near-full-retention accuracy, enabling larger search configurations under fixed device budgets that would otherwise run out of memory.
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.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22102unread
ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling
Woomin Song, Beomjun Kim, Daewon Choi, Sai Muralidhar Jayanthi, Saket Dingliwal, Jinwoo Shin, Aram Galstyan · 2026-05-23
arXiv:2605. 22102v1 Announce Type: new Abstract: A common failure mode in long-horizon agentic test-time scaling is error propagation, where factual errors or invalid deductions introduced at intermediate steps persist in the agent's belief state and contaminate later reasoning.
Read next because ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, correct, soft, eval, line, rate, control. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22102v1 Announce Type: new Abstract: A common failure mode in long-horizon agentic test-time scaling is error propagation, where factual errors or invalid deductions introduced at intermediate steps persist in the agent's belief state and contaminate later reasoning. Existing test-time scaling methods provide limited control over this process, as they often rely on agents to detect their own mistakes, select among flawed trajectories, or refine solutions only after errors have already shaped the reasoning path. We propose ExComm, a communication protocol for exploration-stage agentic test-time scaling. ExComm is motivated by the empirical observation that the majority of intermediate errors in parallel agentic reasoning produce detectable cross-agent factual conflicts. Leveraging the iterative structure of agentic workflows, ExComm periodically audits agent belief states to detect such conflicts, resolves them through a dedicated tool-based verification loop, and returns concise, targeted feedback to the involved agents. Corrections are incorporated through soft belief updates, which append verified feedback rather than overwriting existing beliefs. Furthermore, to prevent collapsing trajectory diversity due to communication, ExComm further introduces a trajectory diversification module that redirects redundant trajectories toward orthogonal strategies. Experiments on AIME 2024, AIME 2025, and GAIA with Gemini-2.5-Flash-Lite and Qwen3.5-4B show that ExComm consistently outperforms strong test-time scaling baselines, achieving average performance gains of 5.7% and 5.0% over the best-performing baselines, respectively. Further analyses demonstrate improved error recovery, favorable scaling behavior, stronger diversity than adapted communication baselines, and the best performance-cost trade-off among the evaluated methods.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22100unread
MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing
Bangbang Zhou, Hangdi Xing, Yifan Chen, Jianjun Xu, Qi Zheng, Feiyu Gao, Zhibo Yang, Shuai Bai, Ming Yan, Jieping Ye, Hongtao Xie · 2026-05-23
arXiv:2605. 22100v1 Announce Type: new Abstract: Document parsing converts visually rich documents into machine-readable structured representations, forming a crucial foundation for information systems.
Read next because MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, extraction, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22100v1 Announce Type: new Abstract: Document parsing converts visually rich documents into machine-readable structured representations, forming a crucial foundation for information systems. Although many benchmarks have been proposed for document parsing, they remain inadequate for realistic scenarios. Existing benchmarks either focus on specific tasks or assess only single-page, text-centric settings, making them insufficient for practical multi-page parsing. Moreover, they lack fine-grained evaluation of semantic continuity, hierarchical structure recovery, and visual content preservation. To address these gaps, we propose MPDocBench-Parse, a benchmark for multi-page document parsing in real-world applications. It contains 433 manually annotated documents with 3,246 pages, covering 15 document types in English and Chinese, with diverse layout styles, and supports document-level end-to-end evaluation. We further design a comprehensive protocol for content fidelity and logical structure, covering text, table, and formula recognition, truncated text and table merging, figure extraction, reading order, and heading hierarchy recovery. Experiments show that, while existing models perform well on basic text extraction, they still suffer clear limitations in semantic continuity integration, visual content parsing, and hierarchical structure recovery. MPDocBench-Parse provides a unified foundation for advancing document parsing toward more realistic scenarios.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, evaluation, benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22090unread
A Camera-Cooperative ISAC Framework for Multimodal Non-Cooperative UAVs Sensing
Wenfeng Wu, Luping Xiang, Kun Yang · 2026-05-23
arXiv:2605. 22090v1 Announce Type: new Abstract: The detection of non-cooperative unmanned aerial vehicles (UAVs) presents significant challenges for Integrated Sensing and Communication (ISAC) systems due to the inherent limitations of single-modal perception and the competition for shared communication and sensing resources.
Read next because A Camera-Cooperative ISAC Framework for Multimodal Non-Cooperative UAVs Sensing 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 "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: alignment, eval, source, rate, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22090v1 Announce Type: new Abstract: The detection of non-cooperative unmanned aerial vehicles (UAVs) presents significant challenges for Integrated Sensing and Communication (ISAC) systems due to the inherent limitations of single-modal perception and the competition for shared communication and sensing resources. To address these challenges, this paper proposes a novel Camera-Cooperative ISAC (CC-ISAC) framework that employs multimodal sensing to enable efficient UAV beam steering and tracking. The proposed framework employs cameras for coarse-grained airspace monitoring and utilizes ISAC for fine-grained, high-precision sensing, forming a complementary perception loop that enhances both sensing accuracy and resource efficiency. Within this framework, two key modules are developed: (1) a Vision-to-Echo Data Alignment (V2EDA) model that aligns visual and echo-domain features through cross-attention mechanisms, and (2) a Multimodal Fusion-Based Estimation (MMFE) model that integrates historical multimodal data with current observations for robust state estimation. Extensive evaluations conducted on the DeepSense 6G dataset demonstrate that the proposed framework achieves an average reduction of 71% in beam steering overhead and 1.69-11.15% in tracking overhead while maintaining high angular estimation accuracy. The CC-ISAC framework effectively mitigates resource contention between sensing and communication, enabling reliable UAV surveillance while freeing substantial system resources for additional communication tasks, thereby representing a practical advancement in ISAC system design.
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 limitation, limitations, evaluation.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22078unread
Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding
Bingjun Luo, Tony Wang, Hanqi Chen, Xinpeng Ding · 2026-05-23
arXiv:2605. 22078v1 Announce Type: new Abstract: Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens while preserving spatiotemporal interactions.
Read next because Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, token, rate, without, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22078v1 Announce Type: new Abstract: Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens while preserving spatiotemporal interactions. Existing methods, such as LLaVA family, utilize simplistic pooling or interpolation techniques that overlook the intricate dynamics of visual tokens. To bridge this gap, we propose ST-GridPool, a novel training-free visual token enhancement method designed specifically for Video LLMs. Our approach integrates Pyramid Temporal Gridding (PTG), which captures multi-grained spatiotemporal interactions through hierarchical temporal gridding, and Norm-based Spatial Pooling (NSP), which preserves high-information visual regions by leveraging the correlation between token norms and semantic richness. Extensive experiments on various benchmarks demonstrate that ST-GridPool consistently enhances performance of Video LLMs without requiring costly retraining. Our method offers an efficient and plug-and-play solution for improving visual token representations. Our code is available in https://github.com/bingjunluo/ST-GridPool.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.22047unread
Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support
Chen Zhan, Xihe Qiu, Xiaoyu Tan, Xibing Zhuang, Gengchen Ma, Yue Zhang, Shuo Li, Peifeng Liu, Xiaoxiao Ge, Liang Liu, Lu Gan · 2026-05-23
arXiv:2605. 22047v1 Announce Type: new Abstract: Large language models perform well on static medical examinations, yet clinical diagnosis often requires iterative evidence gathering under uncertainty.
Read next because Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, control, full, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.22047v1 Announce Type: new Abstract: Large language models perform well on static medical examinations, yet clinical diagnosis often requires iterative evidence gathering under uncertainty. Building on prior interactive evaluation efforts, we introduce an OSCE-inspired standardized patient simulator and a controlled, reproducible benchmark for active diagnostic inquiry. Across 468 cases and 15 models in our protocol, we observe that multi-turn evidence seeking reduces diagnostic accuracy by 12.75% and lowers supporting-evidence quality by 24.36% relative to full-context evaluation; error analyses associate these drops with premature diagnostic closure and inefficient questioning. Together, these results suggest that static full-context benchmarks may overestimate performance in interactive evidence-seeking settings, motivating complementary interactive assessment for safer clinical decision support.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21993unread
ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking
Miaobo Hu, Shuhao Hu, BoKun Wang, Yina Sa, Xin Wang, Xiaobo Guo, Daren Zha, Jun Xiao · 2026-05-23
arXiv:2605. 21993v1 Announce Type: new Abstract: Ranking systems used in decision-support settings should not only order candidates but also expose evidence that can be independently checked.
Read next because ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, alignment, eval, extraction, compare, alone, trained. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21993v1 Announce Type: new Abstract: Ranking systems used in decision-support settings should not only order candidates but also expose evidence that can be independently checked. We study evidence-certified candidate ranking: given an intent_id, a predefined plan skeleton, a window-local candidate roster, and text-derived candidate trajectories with span provenance, a system must output a Top-K list together with doc_id:span evidence certificates whose cited spans are sufficient to recover the decision. We instantiate this task on MAVEN-ERE and RAMS with fixed upstream extraction, window-local randomized candidate identifiers, skeleton-aligned trajectory supervision, hard negatives, and audit references. We introduce Evidence-Coupled Policy Optimization (ECPO), a listwise policy-optimization objective whose action is the joint object of ranking and evidence certificate. ECPO first learns an interpretable trajectory reward from skeleton alignment, argument consistency, and optional graph features; it then optimizes a constrained policy with three coupled rewards: listwise ranking utility, span-level certificate validity, and an evidence-cycle reward computed by a label-free deterministic verifier that reconstructs candidate support from claim-stripped cited spans. This reframes the goal from maximizing ordinary NDCG alone to maximizing CertNDCG and decision-evidence coupling. The evaluation compares ECPO against zero-shot, SFT, and GRPO policies, RM-only scoring with deterministic evidence attachment, grammar/JSON-constrained decoding, validator retry, best-of-N RM selection, and post-hoc evidence rationalization under closed-roster, predicted-roster, and hybrid-roster settings.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses negative, evaluation.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21984unread
Echo: Learning from Experience Data via User-Driven Refinement
Hande Dong, Xiaoyun Liang, Jiarui Yu, Jiayi Lin, Changqing Ai, Feng Liu, Wenjun Zhang, Rongbi Wei, Chaofan Zhu, Linjie Che, Feng Wu, Xin Shen, Dexu Kong, Xiaotian Wang, Qiuyuan Chen, Bingxu An, Yueting Lei, Qiang Lin · 2026-05-23
arXiv:2605. 21984v1 Announce Type: new Abstract: Static "human data" faces inherent limitations: it is expensive to scale and bounded by the knowledge of its creators.
Read next because Echo: Learning from Experience Data via User-Driven Refinement overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rect, source, line, rate, completion, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21984v1 Announce Type: new Abstract: Static "human data" faces inherent limitations: it is expensive to scale and bounded by the knowledge of its creators. Continuous learning from "experience data" - interactions between agents and their environments - promises to transcend these barriers. Today, the widespread deployment of AI agents grants us low-cost access to massive streams of such real-world experience. However, raw interaction logs are inherently noisy, filled with trial-and-error and low information density, rendering them inefficient for direct model training. We introduce Echo, a generalized framework designed to operationalize the transition from raw experience to learnable knowledge, effectively "echoing" environmental feedback back into the training loop for model optimization. In today's agent ecosystem, user refinement serves as a primary source of such feedback: driven by responsibility for the outcome, users rigorously transform flawed agent proposals into verified solutions. These user-driven refinement sequences inherently distill agents' crude attempts into high-quality training signals. Echo systematically harvests these signals to continuously align the agent with real-world needs. Large-scale validation in a production code completion environment confirms that Echo effectively harnesses this pipeline, breaking the static performance ceiling by increasing the acceptance rate from 25.7% to 35.7%.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21974unread
Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables
Jingxuan Qi, Zhiqiang Ye, Yuxiang Feng · 2026-05-23
arXiv:2605. 21974v1 Announce Type: new Abstract: An extraction schema should not reduce knowledge graph fidelity.
Read next because Format-Constraint Coupling in Knowledge Graph Construction from Statistical Tables overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, eval, token, line, extraction, control, trained. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21974v1 Announce Type: new Abstract: An extraction schema should not reduce knowledge graph fidelity. On statistical CSV, however, it can. We study country-by-year time-series matrices, a common layout on open-data portals. In this setting, serialization format and schema constraints interact super-additively. Their joint effect exceeds the sum of independent effects by up to +1.180 (2x2 factorial, 6 datasets). Bootstrap 95% CIs are strictly positive on 4/6 datasets, with strongest evidence on wide Type-II matrices. More critically, a schema applied to a mismatched format can trigger catastrophic mismatch. Fact coverage falls below the unconstrained baseline on 4/6 datasets through entity inflation or extraction refusal. We call this observed pattern format-constraint coupling. Probing and token ablation support a surface-form anchoring explanation centred on column-name references. Controlled variants across format-schema pairings, GraphRAG hosts, and LLM families show the same direction within the measured scope; one LLM family shows only partial activation. The observation also has a diagnostic consequence. Three standard retrieval modes largely mask construction quality (delta <= 1pp), whereas direct graph access exposes gaps up to +47.6pp (p < 0.0001). To support fidelity-aware evaluation, we release CSVFidelity-Bench. It contains 15 datasets, 11 Type-II matrices, 4 Type-III tables, and 1,892 Gold Standard facts across 6 domains.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21962unread
AI-Enabled Serious Games: Integrating Intelligence and Adaptivity in Training Systems
Priyamvada Tripathi, Bill Kapralos · 2026-05-23
arXiv:2605. 21962v1 Announce Type: new Abstract: Serious games are widely used for learning and training across domains such as healthcare, defense, and education.
Read next because AI-Enabled Serious Games: Integrating Intelligence and Adaptivity in Training Systems overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, implement, control, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21962v1 Announce Type: new Abstract: Serious games are widely used for learning and training across domains such as healthcare, defense, and education. Persistent challenges remain, however, including static scenario design, authoring bottlenecks, limited learner modeling, and difficulty implementing meaningful real-time instructional adaptation. Recent advances in artificial intelligence (AI) introduce novel capabilities such as dynamic scenario variation, contextual feedback, adaptive pacing, and learner-state modeling that may help address some of these limitations. At the same time, integrating AI into serious games raises important questions related to validity, transparency, system control, and learner trust. This chapter examines how contemporary AI approaches may support real-time instructional adaptation in serious games. It distinguishes between instructional intelligence, defined as a system's capacity to infer learner knowledge and reason about pedagogically appropriate responses, and adaptivity, defined as the ability to modify instructional actions during interaction. A historical synthesis of adaptive learning systems is presented, tracing developments from early computer-assisted instruction through intelligent tutoring systems (ITS), dynamic difficulty adjustment (DDA), authoring platforms, learning analytics, and recent AI-enabled architectures. Building on this perspective, the chapter discusses how large language models (LLMs), reinforcement learning (RL), and agent-based architectures may contribute to more integrated forms of intelligence and adaptivity in serious games. It also highlights practical and research challenges associated with AI-enabled systems, including explainability, validation, computational cost, and the limited empirical evidence regarding long-term learning outcomes in AI-enabled serious games.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21902unread
Planning in the LLM Era: Building for Reliability and Efficiency
Michael Katz, Harsha Kokel, Kavitha Srinivas, Shirin Sohrabi · 2026-05-23
arXiv:2605. 21902v1 Announce Type: new Abstract: Growing attention to intelligent agents has put a spotlight on one of their central capabilities: planning.
Read next because Planning in the LLM Era: Building for Reliability and Efficiency 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 "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: alignment, source, line, without, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21902v1 Announce Type: new Abstract: Growing attention to intelligent agents has put a spotlight on one of their central capabilities: planning. Early attempts to leverage large language models (LLMs) for planning relied on single-shot plan generation, followed by hybrid approaches that coupled LLMs with limited external search. These methods, unsound and incomplete by their very nature, often require substantial resources without yielding better solutions on unseen problems. As the limitations of LLMs become clearer, recent work has shifted toward using them at solution construction time -- generating symbolic solvers for a family of problems that can be verified and then used efficiently at inference time. This trend reflects the growing need for agents that are both reliable and resource-efficient. It also offers a path towards generating maintainable planners with minimal dependence on language models at inference time. In this paper, we argue that this shift reflects a broader realignment of the planning field in the LLM era. We examine three major categories of planner-generation methods, discuss their current limitations, and outline research steps towards a more reliable and efficient LLM-based generation of planners.
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 limitation, limitations.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21822unread
Implicit Safety Alignment from Crowd Preferences
Qian Lin, Daniel S. Brown · 2026-05-23
arXiv:2605. 21822v1 Announce Type: new Abstract: Reinforcement Learning from Human Feedback (RLHF) can reveal implicit objectives such as safety considerations that go beyond task completion.
Read next because Implicit Safety Alignment from Crowd Preferences 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, alignment, rate, without, trained, completion, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21822v1 Announce Type: new Abstract: Reinforcement Learning from Human Feedback (RLHF) can reveal implicit objectives such as safety considerations that go beyond task completion. In this work, we focus on the common safety criteria embedded in crowd preference datasets, where different users may express distinct preferences or objectives, yet follow similar safety principles. Our aim is to discover shared safety criteria from crowd preferences and then transfer them to downstream RL tasks to regularize agent behavior and enforce safety. We first show that direct reward combination-optimizing a preference-learned reward model together with downstream task rewards-has inherent limitations. Motivated by this, we propose Safe Crowd Preference-based RL, a hierarchical framework that extracts safety-aligned skills from crowd preferences and composes them via a high-level policy to safely solve downstream tasks. Experiments across safe RL environments and a preliminary LLM-style task with diverse user goals and shared safety constraints demonstrate that our approach substantially lowers safety costs without access to explicit safety rewards, while achieving task performance comparable to oracle methods trained with ground-truth safety signals.
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.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21810unread
Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents
Zijian Du, Nathaniel Pinckney · 2026-05-23
arXiv:2605. 21810v1 Announce Type: new Abstract: Complex Verilog Design Problems (CVDP) challenge hardware LLM agents because solving them requires localizing verifier-relevant RTL, testbenches, include paths, and build dependencies inside large repository snapshots, making precise edits, and recovering from sparse hidden-verifier failures.
Read next because Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, without, candidate, test, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21810v1 Announce Type: new Abstract: Complex Verilog Design Problems (CVDP) challenge hardware LLM agents because solving them requires localizing verifier-relevant RTL, testbenches, include paths, and build dependencies inside large repository snapshots, making precise edits, and recovering from sparse hidden-verifier failures. We present Trace2Skill, a test-time scaling framework that improves a hardware agent without RTL-specialized model fine-tuning. Rather than training a new model or only sampling more candidate solutions, Trace2Skill treats the agent's natural-language skill as an evolvable policy. It mines repeated rollout traces for success and failure modes, converts them into dense diagnostics and oracle lessons, and uses an oracle, mutator, and selector loop to produce task-specific skills that guide later search, editing, validation, and recovery. Because final pass/fail labels are often too coarse for hard failures, Trace2Skill also supports bounded runtime dense verifier feedback that returns sanitized functional observations while keeping hidden harnesses and reference solutions inaccessible to the agent. This feedback helps guide skill evolution and agent execution by connecting skill text, verifier evidence, and downstream behavior. Across hard CVDP tasks that defeat the seed CVDP agent, including tasks that also defeat frontier coding agents, Trace2Skill with dense verifier feedback substantially improves task pass rates and produces breakthrough passes on previously unsolved tasks, without requiring high-quality fine-tuning data, specialized RTL model training, or model weight updates. The same framework provides a general test-time scaling strategy that can extend beyond digital design to other verifiable EDA tasks.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, failures.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21778unread
What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct
Meryl Ye, Lujain Ibrahim, Jessica Y. Bo, Myra Cheng, Ida Mattsson, Daniel Vennemeyer, Robert Kraut, Steve Rathje · 2026-05-23
arXiv:2605. 21778v1 Announce Type: new Abstract: AI sycophancy has become a prominent concern in large language model (LLM) research.
Read next because What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (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: persona, rect, under, correct, eval, rate, compare, position. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21778v1 Announce Type: new Abstract: AI sycophancy has become a prominent concern in large language model (LLM) research. Yet the term lacks a consistent definition and has been applied to behaviors ranging from agreeing with a user's false claim to excessively praising the user to withholding corrective feedback. When researchers, companies, and policymakers use the same term to describe different behaviors, evaluation results become difficult to compare, mitigation strategies fail to transfer, and systems that are resistant to one form of sycophancy continue exhibiting other forms. To address this, we make two contributions. First, we reviewed 70 papers on AI sycophancy to develop a taxonomy of how the behavior has been defined and measured. The taxonomy distinguishes (1) whether a model is sycophantic toward a user's positions and beliefs, or toward the user's broader personal traits and emotions, and (2) whether this occurs through explicit, direct language or more implicit, subtle behaviors such as framing, omission, or tone. Mapping existing literature to our taxonomy reveals that current research has focused on overt forms of sycophancy toward users' beliefs, leaving more subtle and person-directed behaviors relatively understudied. Second, we surveyed 106 experts in AI sycophancy and related fields to examine whether researchers agree on which model behaviors are sycophantic. While experts are nearly unanimous in believing that sycophancy is a significant problem in current AI systems (94.3% agree), they disagree substantially on which specific behaviors qualify. Together, these findings demonstrate that AI sycophancy is a broad family of behaviors with different measurement challenges, intervention requirements, and governance implications. Our taxonomy provides a shared vocabulary for understanding and addressing these behaviors.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21758unread
A Causal Argumentation Method for Explainability of Machine Learning Models
Henry Salgado, Meagan R. Kendall, Martine Ceberio · 2026-05-23
arXiv:2605. 21758v1 Announce Type: new Abstract: Explainable AI (XAI) methods identify which features are relevant to a model's predictions but often fail to clarify why certain decisions are made.
Read next because A Causal Argumentation Method for Explainability of Machine Learning Models 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, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21758v1 Announce Type: new Abstract: Explainable AI (XAI) methods identify which features are relevant to a model's predictions but often fail to clarify why certain decisions are made. In this work, we present a novel method that integrates causality with argument-based reasoning to explain why models may be making predictions. Our approach first identifies causal relationships among variables using causal discovery methods and then translates these into a Bipolar Argumentation Framework (BAF) to represent supportive and opposing interactions among features. By using semi-stable semantics, we find extensions of features that explain why certain outcomes may have been chosen. We demonstrate our method on two benchmark datasets and compare its results against standard post-hoc explainability approaches.
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 benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21743unread
Who Uses AI? Platforms, Workforce, and AI Exposure
Michelle Yin, Burhan Ogut · 2026-05-23
arXiv:2605. 21743v1 Announce Type: new Abstract: A growing literature uses artificial intelligence platform conversation logs to measure occupation exposure.
Read next because Who Uses AI? Platforms, Workforce, and AI Exposure 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, control, factor. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21743v1 Announce Type: new Abstract: A growing literature uses artificial intelligence platform conversation logs to measure occupation exposure. We show that these scores partly measure platform user base rather than the workforce. Holding outcome, sample, controls, and estimator fixed while varying only the platform input changes the post-ChatGPT employment coefficient by a factor of 1.9, and within-vendor consumer-versus-enterprise channels produce estimates that disagree in sign. Reweighting to Bureau of Labor Statistics workforce shares attenuates estimates by 42 to 93 percent. We formalize the non-classical measurement error, derive probability limits and partial-identification bounds for employment elasticities. The bias understates substitution more than augmentation.
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 bias.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21740unread
SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?
Kevin Han, Renfei Zhang, Kathy Wei, Hamed Mahdavi, Niloofar Mireshghallah, Amir Farimani · 2026-05-23
arXiv:2605. 21740v1 Announce Type: new Abstract: LLM agents have incredible potential for scientific discovery applications.
Read next because SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks? overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, source, rate, full, test. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21740v1 Announce Type: new Abstract: LLM agents have incredible potential for scientific discovery applications. However, the performance of LLM agents on real-world, small molecule drug design (SMDD) tasks across diverse chemistries and targets is unclear. Current evaluation methods are either ad hoc, too simple for real-world discovery, limited in scale, or restricted to single-turn question answering. In effort to standardize the evaluation of LLM agents on small molecule design, we introduce SMDD-Bench, a challenging, multi-turn, long-horizon agentic benchmark consisting of 502 guaranteed-solvable task instances spanning 5 task types: 2D Pharmacophore Identification, Interaction Point Discovery, Scaffold Hopping, Lead Optimization, and Fragment Assembly. SMDD-Bench tasks span a wide region of chemical space and involve 102 unique protein targets. Completely solving the benchmark would require having strong chemical and biological reasoning and 3D intuition, understanding specialized tool use, and displaying planning expertise over a limited number of oracle calls. We benchmark 7 frontier open and closed source LLMs and find even the most performant LLM, GPT5.4, solves only 40.2\% of tasks. We hope SMDD-Bench provides a standardized testbed to invigorate the field towards training and evaluating LLM agents for fully autonomous computational drug design. We host a public leaderboard at smddbench.com .
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21739unread
AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence
Kate M. Lubrano, Faisal Sayed, Ankita Rathod, Akshansh, Craver Corbyn Thomas-Smith, Mark E. Whiting, Karina Nguyen · 2026-05-23
arXiv:2605. 21739v1 Announce Type: new Abstract: Emotional intelligence (EI), the ability to perceive, understand, and respond appropriately to others' emotional states, is central to human communication, and increasingly important to assess as LLMs assume conversational roles in everyday life.
Read next because AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligence overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rect, under, alignment, eval, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21739v1 Announce Type: new Abstract: Emotional intelligence (EI), the ability to perceive, understand, and respond appropriately to others' emotional states, is central to human communication, and increasingly important to assess as LLMs assume conversational roles in everyday life. Existing EI benchmarks rely on synthetic prompts, single-turn cases, or third-party annotation. These approaches do not directly measure how models infer and respond to a participant's emotional state over the course of a real conversation. We introduce AttuneBench, a benchmark grounded in 200 genuine multi-turn human-model conversations in which participants conversed with anonymized LLMs and provided turn-by-turn annotations of their emotional state, the model's behavior, and their preferred responses. Across 11 evaluated models, we find that model rankings on emotion recognition, behavioral classification, preference prediction, and judged response quality are largely independent, indicating that emotionally intelligent behavior decomposes into separable capabilities. Preference alignment and response-quality judgments are substantially more model-discriminating than emotion-label accuracy. These results indicate that emotionally intelligent behavior requires predicting what kind of response a specific user wants in context, a distinction that aggregate scoring can obscure and that single-turn or synthetic formats cannot directly capture across turns. AttuneBench provides a framework for assessing each of these capabilities and for diagnosing model-specific strengths and failure modes in emotionally salient conversation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21706unread
Latent-space Attacks for Refusal Evasion in Language Models
Giorgio Piras, Raffaele Mura, Fabio Brau, Maura Pintor, Luca Oneto, Fabio Roli, Battista Biggio · 2026-05-23
arXiv:2605. 21706v1 Announce Type: new Abstract: Safety-aligned language models are trained to refuse harmful requests, yet refusal behavior can be suppressed by steering their internal representations.
Read next because Latent-space Attacks for Refusal Evasion in Language Models 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: latin, rect, under, line, rate, project, control, trained. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21706v1 Announce Type: new Abstract: Safety-aligned language models are trained to refuse harmful requests, yet refusal behavior can be suppressed by steering their internal representations. Existing methods do so by ablating a refusal direction from model activations, aiming to remove refusal from the model's residual stream. Despite their empirical success, these methods lack a principled account of the latent-space transformation they induce and why it suppresses refusal. In this work, we recast refusal suppression as a latent-space evasion attack against linear probes trained to separate refused from answered prompts. Under this view, prior work's difference-in-means direction naturally defines such a probe, and its ablation is exactly a projection onto its decision boundary, i.e., a minimum-confidence evasion attack. This perspective not only explains the empirical success of prior work but also admits a key limitation: evasion stops at the decision boundary, motivating the need to push representations further into the compliant region, i.e., where the model answers. We leverage this by proposing a Controlled Latent-space Evasion attack that projects representations past the boundary with an optimized confidence. We achieve state-of-the-art attack success rate across 15 instruction-tuned, multimodal, and reasoning models, outperforming existing refusal-ablation baselines and specialized jailbreak attacks.
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.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21645unread
AOP-Wiki EMOD 3.0: Data Model Expansions and Content Evaluation Framework for Using Agentic AI to Improve Integration between AOPs and New Approach Methodologies (NAMs)
Virginia K. Hench, J. Harry Caufield, Sierra A. T. Moxon, Jason M. O'Brien, Stephen W. Edwards · 2026-05-23
arXiv:2605. 21645v1 Announce Type: new Abstract: Adverse Outcome Pathways (AOP) are logic models that causally link biological mechanisms that can be measured in a lab to adverse outcomes, relevant to chemical regulatory endpoints.
Read next because AOP-Wiki EMOD 3.0: Data Model Expansions and Content Evaluation Framework for Using Agentic AI to Improve Integration between AOPs and New Approach Methodologies (NAMs) overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, contexts, test, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21645v1 Announce Type: new Abstract: Adverse Outcome Pathways (AOP) are logic models that causally link biological mechanisms that can be measured in a lab to adverse outcomes, relevant to chemical regulatory endpoints. AOPs contextualize new approach methodologies (NAMs), in vitro and in silico methods used as alternatives to animal testing and the sequential events in an AOP serve as multi-scale models spanning biological scales. The AOP-Wiki serves as the global repository for AOPs. While the AOP-Wiki has played a central role in AOP expansion over the past decade, constraints within the current data model and application infrastructure limit the AOP-Wiki from supporting continued AOP growth and evolution. Yet, the transformative power of agentic AI has re-invigorated AOP-Wiki data modernization efforts at a time when core AOP principles can be harnessed to inform use of AI for aggregating and structuring AOP-relevant information. Seizing upon this momentum, we present AOP-Wiki EMOD 3.0, the third in a series of evidence model prototypes, which concretely demonstrates data model expansions and our vision for how the AOP-Wiki might be transformed to better serve regulatory science and emergent use of AOPs in biomedical and One Health contexts. We aim to lay a foundation to support computationally-generated AOPs and quantitative AOPs (qAOPs) by focussing on solutions for AOP-Wiki internal quality improvement, evidence structuring to enhance AOP FAIRness and AI-readiness, and improved integration between the AOP framework and NAMs to better serve next generation risk assessment.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21630unread
MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis
Haiyang Shen, Taian Guo, Xuanzhong Chen, Mugeng Liu, Weichen Bi, Wenchun Jing, Sixiong Xie, Zhuofan Shi, Yudong Han, Chongyang Pan, Siqi Zhong, Jinsheng Huang, Ming Zhang, Yun Ma · 2026-05-23
arXiv:2605. 21630v1 Announce Type: new Abstract: Although LLMs have made substantial progress in reasoning, systematically producing frontier-level reasoning data remains difficult.
Read next because MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis 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, source, line, rate, implement, control. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21630v1 Announce Type: new Abstract: Although LLMs have made substantial progress in reasoning, systematically producing frontier-level reasoning data remains difficult. Existing synthesis methods often have limited visibility into the structural factors that govern problem difficulty, which can result in narrow diversity and unstable difficulty control. In this work, we view the difficulty of a reasoning problem as arising from the accumulation of atomic knowledge-reasoning transformations, which we term thought modes. Building on this perspective, we propose MindLoom, a framework for synthesizing frontier-level reasoning data through compositional thought mode engineering. Given a collection of hard problems with verified solutions, MindLoom first decomposes those solutions into thought mode chains that reveal each problem's construction logic. It then trains a retrieval model that matches problem states to compatible thought modes, providing guidance on which reasoning challenges to introduce during synthesis. New problems are composed by iteratively applying retrieved thought modes to seed questions, with distribution-aligned sampling to encourage diverse reasoning coverage. Finally, a rollout-based judging stage labels generated questions by difficulty and supplies judged-correct responses for supervised fine-tuning. We evaluate MindLoom on nine benchmarks covering five STEM disciplines and four mathematical reasoning tasks across multiple model families and sizes. Models fine-tuned on MindLoom-generated data achieves favorable performances over base models, distillation, and external-data baselines across the reported benchmarks. Ablation studies indicate the contribution of each component, and further analysis suggests that MindLoom covers a broad range of reasoning patterns while maintaining useful difficulty control. We have open-sourced our implementation at https://github.com/EachSheep/MindLoom.
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.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21622unread
TO-Agents: A Multi-Agent AI Pipeline for Preference-Guided Topology Optimization
Isabella A. Stewart, Hongrui Chen, Faez Ahmed · 2026-05-23
arXiv:2605. 21622v1 Announce Type: new Abstract: Topology optimization can generate efficient structures, but designers often must manually translate qualitative intent, such as desired visual style, product experience, or manufacturability into solver settings that are not directly tied to those preferences.
Read next because TO-Agents: A Multi-Agent AI Pipeline for Preference-Guided Topology Optimization 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, without, lora, language. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21622v1 Announce Type: new Abstract: Topology optimization can generate efficient structures, but designers often must manually translate qualitative intent, such as desired visual style, product experience, or manufacturability into solver settings that are not directly tied to those preferences. We present TO-Agents, a multi-agent AI framework that connects natural-language design intent with iterative topology optimization. The framework converts a human-provided problem description into validated solver inputs, runs a topology optimization solver, renders the resulting 3D topology, and uses multi-view vision-language reasoning with an independent judge agent to critique each result and revise solver parameters. We evaluate the framework on two long-horizon design tasks: a cantilever beam benchmark and a phone-stand product design. In both tasks, the designer specifies an aesthetic preference for hierarchically branched structures inspired by natural tree morphologies, and the system performs four revision cycles across ten independent replicates. TO-Agents produces at least one preference-aligned design in 60% of trials for each case study, corresponding to up to 6x more successful trials than an ablated pipeline without visual or historical feedback. Judge scores and human evaluations show that the pipeline can identify effective parameter levers, recover from poor revisions, and expand design exploration. A manufacturing agent further post-processes top-ranked designs for additive manufacturing, enabling end-to-end intent-to-prototype design. We also identify failure modes, including overshooting, selective memory, misplaced tools, and incorrect parameter reasoning. These results suggest that agentic topology optimization can shift designers from low-level parameter tuning toward higher-level specification of form and function, while highlighting safeguards needed for reliable autonomous engineering design.
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, evaluation, benchmark.
- score 100arxiv cs.AI (Artificial Intelligence)arxiv:2605.21602unread
Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs
Dylan Feng, Pragya Srivastava, Cassidy Laidlaw · 2026-05-23
arXiv:2605. 21602v1 Announce Type: new Abstract: Many safety and alignment failures of large language models (LLMs) occur due to out-of-distribution (OOD) situations: unusual prompt or response patterns that are unforeseen by model developers.
Read next because Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs 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, line, trained, test, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv:2605.21602v1 Announce Type: new Abstract: Many safety and alignment failures of large language models (LLMs) occur due to out-of-distribution (OOD) situations: unusual prompt or response patterns that are unforeseen by model developers. We systematically study whether LLM monitoring pipelines can detect these OOD alignment failures by introducing a benchmark called Misalignment Out Of Distribution (MOOD). It is difficult to find failures that are truly OOD for off-the-shelf models trained on vast safety datasets. We sidestep this by including a restricted training set in MOOD that we use to train our own monitors, as well as seven test sets with diverse alignment failures that are outside the training distribution. Using MOOD, we find that guard models (safety classifiers) often fail to generalize OOD. To fix this, we propose combining guard models with OOD detectors. We test four types of OOD detectors and find that a combination of a guard model with Mahalanobis distance and perplexity-based OOD detectors can improve recall from 39% to 45%. We also establish positive scaling trends across model scales for monitors that combine a guard model and OOD detector; we find that incorporating OOD detection into monitoring achieves a higher recall gain than using a guard model with 20 times more parameters. Our work suggests that OOD detection should be a crucial component of LLM monitoring and provides a foundation for further work on this important problem.
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, failures, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21646unread
Alike Parts: A Feature-Informed Approach to Local and Global Prototype Explanations
Jacek Karolczak, Jerzy Stefanowski · 2026-05-23
arXiv:2605. 21646v1 Announce Type: new Abstract: Prototype-based explanations offer an intuitive, example-based approach to support the interpretability of machine learning black box classifiers but often lack feature-level granularity.
Read next because Alike Parts: A Feature-Informed Approach to Local and Global Prototype Explanations overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rate, does, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21646v1 Announce Type: new Abstract: Prototype-based explanations offer an intuitive, example-based approach to support the interpretability of machine learning black box classifiers but often lack feature-level granularity. We introduce a framework that integrates feature importance at two levels to address this gap. First, for local explanations, we propose \textit{alike parts}: a method that uses feature importance scores to highlight the most relevant, shared feature subsets between a classified instance and its nearest prototype, guiding user attention. Second, we augment the global prototype selection objective function with a feature importance term to actively promote diversity in the feature attributions of the selected prototypes. Experiments on six benchmark datasets show that this augmented selection process maintains or, in some cases, increases the prediction fidelity of the surrogate model, suggesting that feature diversity does not compromise model fidelity.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21610unread
AgForce Enables Antigen-conditioned Generative Antibody Design
Mansoor Ahmed, Murray Patterson · 2026-05-23
arXiv:2605. 21610v1 Announce Type: new Abstract: Antibody design methods condition on antigen structure to generate complementarity-determining regions (CDR), yet a systematic evaluation of baseline methods reveals that they largely ignore the antigen input.
Read next because AgForce Enables Antigen-conditioned Generative Antibody Design overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, eval, source, line, rate, binding. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21610v1 Announce Type: new Abstract: Antibody design methods condition on antigen structure to generate complementarity-determining regions (CDR), yet a systematic evaluation of baseline methods reveals that they largely ignore the antigen input. We identify three failure modes that explain this behavior. Antigen blindness arises because models derive predictions from antibody framework context rather than antigen information, producing nearly identical CDRs regardless of the target. Vocabulary collapse reduces predicted amino acids to three to five per position, far below the ground truth distribution in native sequences. Moreover, any model trained with standard per-position cross-entropy converges to the positional marginal distribution, making it provably unable to produce antigen-specific sequence predictions. We propose a novel encoder-decoder architecture called AgForce, that uses a graph neural network (GNN) as the encoder and specialized decoders for sequence-structure co-design. Specifically, we apply framework dropout, gated bottlenecks, and hyperbolic cross attention that prevent the antibody shortcut path. In the decoder, a Mixture Density Network (MDN) sequence head with Potts-like pairwise coupling and annealed Multiple Choice Learning (aMCL) replaces the cross-entropy objective with a multi-component distribution whose optimal solution differs from the positional marginal. An antigen cycle consistency head routes gradients through the sequence decoder, forcing predicted distributions to encode antigen identity. AgForce achieves the best binding quality and sequence recovery simultaneously on the CHIMERA-Bench dataset, improving amino acid recovery by 8% over the strongest sequence baseline while surpassing the baselines across all interface metrics, and nearly doubling the effective vocabulary of GNN methods. The source code is available at: https://github.com/mansoor181/ag-force.git
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21606unread
When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
Xiaogeng Liu, Xinyan Wang, Yingzi Ma, Yechao Zhang, Chaowei Xiao · 2026-05-23
arXiv:2605. 21606v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains a student on its own rollouts using a privileged teacher, but its standard objective weights all generated tokens equally, implicitly treating the privileged teacher target as equally reliable at every student-visited prefix.
Read next because When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, latin, rect, under, correct, eval, prefix, token. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21606v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains a student on its own rollouts using a privileged teacher, but its standard objective weights all generated tokens equally, implicitly treating the privileged teacher target as equally reliable at every student-visited prefix. Existing entropy-based OPD methods relax this uniformity by modulating token-level supervision with teacher entropy, but high teacher entropy in reasoning has an ambiguous reliability meaning: it can reflect either non-viable uncertainty or benign solution diversity. To identify this phenomenon, we introduce a branch-viability diagnostic. Specifically, we record next-token alternatives from the privileged-answer teacher prompt, force each alternative after the student prompt plus its on-policy spine prefix, and test whether the resulting student-template continuation recovers the correct answer. On Qwen3-4B, we find that an oriented within-sequence position score is the strongest tested predictor of teacher-token reliability, reaching an area-under-ROC-curve (AUROC) of 0.83; local uncertainty scores are at most 0.57. Motivated by this trajectory-level structure, we propose Position-Weighted On-Policy Self-Distillation (PW-OPSD), which applies an increasing position weight while keeping the same student rollout, privileged teacher pass, and clipped forward-KL target as OPSD. In our comprehensive evaluations with different random seeds, the diagnostic-derived PW-OPSD improves AIME 2024 and AIME 2025 Avg@12 by +1.0 and +1.1 points, and a generalization evaluation on two larger-scale models from different families, DeepSeek-R1-Distill-Llama-8B and Olmo-3-7B-Think, also demonstrates consistent aggregate Avg@12 improvements. These results show that teacher-token reliability in reasoning distillation is trajectory-structured and can be utilized without additional teacher computation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21600unread
ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning
Mansoor Ahmed, Spencer VonBank, Nadeem Taj, Sujin Lee, Naila Jan, Murray Patterson · 2026-05-23
arXiv:2605. 21600v1 Announce Type: new Abstract: Computational antibody CDR design methods condition on antigen structure to generate binding loops, yet existing architectures conflate two fundamentally distinct sub-problems: identifying which CDR positions will contact the antigen, and selecting amino acids at those positions.
Read next because ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rate, binding, stage, position, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21600v1 Announce Type: new Abstract: Computational antibody CDR design methods condition on antigen structure to generate binding loops, yet existing architectures conflate two fundamentally distinct sub-problems: identifying which CDR positions will contact the antigen, and selecting amino acids at those positions. This conflation forces models to learn contact reasoning implicitly through uniform message passing, diluting antigen signal across all positions equally. We introduce ConTact, a contact-then-act architecture that explicitly decomposes CDR design into three cascaded stages: learning surface complementarity fingerprints, predicting CDR-antigen contacts, and injecting contact-gated antigen features into the sequence head. A distance-biased cross-attention module encodes geometric priors favoring spatial neighbors, while a contact-weighted cross-entropy loss concentrates gradient signal on binding-critical positions. On CHIMERA-Bench dataset, ConTact achieves the best structural quality (7% RMSD improvement over the next-best baseline), best epitope awareness (10% F1 score over GNN baselines), and competitive sequence recovery (AAR 0.38) among several CDR-H3 design baselines.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21566unread
Calibration, Uncertainty Communication, and Deployment Readiness in CKD Risk Prediction: A Framework Evaluation Study
Michael O. Eniolade · 2026-05-23
arXiv:2605. 21566v1 Announce Type: new Abstract: Machine learning models for chronic kidney disease (CKD) risk prediction often post strong discrimination scores on internal test sets.
Read next because Calibration, Uncertainty Communication, and Deployment Readiness in CKD Risk Prediction: A Framework Evaluation Study overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, under, distributional, eval, rate, without, trained. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21566v1 Announce Type: new Abstract: Machine learning models for chronic kidney disease (CKD) risk prediction often post strong discrimination scores on internal test sets. Calibration and uncertainty quantification get far less attention, leaving clinicians without reliable information about whether the probability outputs are accurate. We trained five classifiers on the UCI CKD dataset (400 patients, 62.5% CKD prevalence): logistic regression, random forest, XGBoost, SVM with Platt scaling, and Gaussian naive Bayes. We evaluated each across calibration quality, conformal prediction coverage, and an eight-criterion deployment readiness framework. A distributional stress-test applied the best-calibrated variant of each model to the open-access MIMIC-IV demo cohort (97 patients, 23.7% CKD) to assess behaviour under prevalence shift and feature missingness. We measured calibration before and after Platt scaling and isotonic regression using Expected Calibration Error and Brier Score, and quantified uncertainty through split conformal prediction targeting 90% marginal coverage. All five models reached AUROC 1.00 on the UCI test set. Isotonic recalibration reduced internal ECE to 0.000-0.022. On MIMIC-IV, AUROC fell to 0.48-0.58, ECE rose to 0.68-0.76, and conformal coverage dropped from 0.80-0.98 to 0.21-0.25 against a 90% target. No model scored above 4 out of 16 on the deployment readiness checklist. Near-perfect internal performance did not transfer. Calibration stability and conformal coverage should be evaluated on external data before any clinical prediction model moves toward deployment.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21565unread
Leveraging Self-Paced Curriculum Learning for Enhanced Modality Balance in Multimodal Conversational Emotion Recognition
Phuong-Anh Nguyen, The-Son Le, Duc-Trong Le, Cam-Van Thi Nguyen · 2026-05-23
arXiv:2605. 21565v1 Announce Type: new Abstract: Multimodal Emotion Recognition in Conversations (MERC) is a crucial task for understanding human interactions, where multimodal approaches integrating language, facial expressions, and vocal tone have achieved significant progress.
Read next because Leveraging Self-Paced Curriculum Learning for Enhanced Modality Balance in Multimodal Conversational Emotion Recognition 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: under, alignment, line, rate, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21565v1 Announce Type: new Abstract: Multimodal Emotion Recognition in Conversations (MERC) is a crucial task for understanding human interactions, where multimodal approaches integrating language, facial expressions, and vocal tone have achieved significant progress. However, modality misalignment and imbalanced learning remain major challenges, limiting the effective utilization of multimodal information. To address this issue, we propose a plug-and-play framework based on Self-Paced Curriculum Learning (SPCL) for MERC. We introduce a dual-level Difficulty Measurer that captures both utterance-level and conversation-level challenges. The utterance-level score models fine-grained modality-specific difficulty, while the conversation-level score captures broader dialogue structures, including emotional dependencies and modality coherence. Based on these scores, the Learning Scheduler dynamically guides training from easier to more difficult instances. By integrating SPCL into existing MERC architectures, our method alleviates modality imbalance and improves model robustness. Extensive experiments on the IEMOCAP and MELD datasets demonstrate consistent improvements across different architectures and modality settings. On IEMOCAP, SPCL improves weighted F1-score by approximately +1.2% to +6.6% over baseline models, while on MELD, gains reach up to +10.4%. These results highlight the effectiveness and generalizability of SPCL as a lightweight plug-and-play module for multimodal emotion recognition.
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 robustness.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21563unread
Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction
Fan Zhang, Simon Deltadahl, Majid Lotfian Delouee, Daniel Kreuter, Joseph Taylor, Allerdien Visser, BloodCounts Consortium, James H. F. Rudd, Nicholas S. Gleadall, Suthesh Sivapalaratnam, Folkert Asselbergs, Martijn C. Schut, Michael Roberts · 2026-05-23
arXiv:2605. 21563v1 Announce Type: new Abstract: Recent reviews find that the vast majority of published healthcare federated learning (FL) studies never reach real-world deployment.
Read next because Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, class, under, eval, line, rate, extraction. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21563v1 Announce Type: new Abstract: Recent reviews find that the vast majority of published healthcare federated learning (FL) studies never reach real-world deployment. We developed an embedding-based FL pipeline for iron deficiency prediction from routine full blood count (FBC) data and deployed it across real institutional environments at Amsterdam University Medical Centre (AUMC) and NHS Blood and Transplant (NHSBT), two clinical environments that differ markedly in iron deficiency prevalence, ferritin distribution, and subject populations. A frozen domain-specific haematology foundation model, DeepCBC, performs site-local representation extraction, restricting federated training to a compact downstream classifier and substantially reducing recurrent communication relative to full-encoder federation. The two clinical datasets are structurally not independent and identically distributed (non-IID), with heterogeneity arising from distinct population differences rather than sampling artefacts. Runtime governance is enforced by FLA$^3$, a healthcare-oriented FL platform providing study-scoped execution, policy-based authorisation, and signed audit logging. Standard sample-size-weighted aggregation (FedAvg) reduced the area under the receiver operating characteristic curve (ROC-AUC) at both sites relative to local-only training, as the global update was biased towards the larger AUMC distribution. FedMAP, a personalised aggregation method, raised ROC-AUC from 0.9470 to 0.9594 at AUMC and from 0.8558 to 0.8671 at NHSBT relative to local-only training, achieving the highest macro ROC-AUC of 0.9133 and the best macro balanced accuracy overall. These results support personalised aggregation in clinical federations where client sample size and task relevance diverge substantially.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21561unread
Objective-Induced Bias and Search Dynamics in Multiobjective Unsupervised Feature Selection
Mathieu Cherpitel, Thomas B\"ack, Martijn R. Tannemaat, Anna V. Kononova · 2026-05-23
arXiv:2605. 21561v1 Announce Type: new Abstract: Unsupervised feature selection is commonly formulated as a multiobjective optimisation problem that jointly optimises subset quality and subset size.
Read next because Objective-Induced Bias and Search Dynamics in Multiobjective Unsupervised Feature Selection overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, eval, rate, compare, control, factor, test. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21561v1 Announce Type: new Abstract: Unsupervised feature selection is commonly formulated as a multiobjective optimisation problem that jointly optimises subset quality and subset size. Yet the behaviour of this formulation depends critically on the choice of evaluation objective, the direction of subset-size regularisation, and the initialisation strategy. We study these factors in a controlled setting using a synthetic dataset with known informative, redundant, and irrelevant feature types. Six formulations are compared by combining three evaluation objectives: accuracy, silhouette score, and PCA reconstruction loss with subset-size minimisation or maximisation. The results show that formulation strongly affects both search dynamics and the quality of the resulting Pareto front. Silhouette-based formulations exhibit a strong bias toward trivial low-cardinality solutions and remain weak proxies for predictive performance. In contrast, the proposed PCA loss objective produces compact subsets with test accuracy comparable to subsets obtained by directly optimising supervised accuracy. Overall, the study shows that objective design is central to effective multiobjective unsupervised feature selection.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses bias, evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21560unread
AutoMCU: Feasibility-First MCU Neural Network Customization via LLM-based Multi-Agent Systems
Penglin Dai, Zijie Zhou, Xincao Xu, Junhua Wang, Xiao Wu, Lixin Duan · 2026-05-23
arXiv:2605. 21560v1 Announce Type: new Abstract: Deploying neural networks on microcontroller units (MCUs) is critical for edge intelligence but remains challenging due to tight memory, storage, and computation constraints.
Read next because AutoMCU: Feasibility-First MCU Neural Network Customization via LLM-based Multi-Agent Systems 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, rate, compare, control, full, chain. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21560v1 Announce Type: new Abstract: Deploying neural networks on microcontroller units (MCUs) is critical for edge intelligence but remains challenging due to tight memory, storage, and computation constraints. Existing approaches, such as model compression and hardware-aware neural architecture search (HW-NAS), often depend on proxy metrics, incur high search cost, and do not fully bridge the gap between architecture design and verified deployment. This paper presents AutoMCU, a feasibility-first large language model (LLM)-based multi-agent system for automated neural network customization under MCU constraints. Given natural-language task requirements and hardware specifications, AutoMCU iteratively generates structured architecture candidates, filters infeasible designs through vendor toolchain feedback before training, evaluates feasible models under a controlled protocol, and verifies deployability through backend-grounded deployment analysis. AutoMCU includes two key mechanisms: 1) hardware-in-the-loop architecture generation for early elimination of undeployable candidates under RAM and Flash constraints, and 2) state-isolated multi-agent scheduling for stable coordination of proposal, training, evaluation, and deployment stages. Experiments on CIFAR-10 and CIFAR-100 under strict MCU constraints show that AutoMCU achieves competitive accuracy while reducing customization time to about 1--2 hours, compared with hundreds of GPU hours for representative MCU-oriented HW-NAS baselines. Comparisons with ColabNAS and the LLM-based NAS method GENIUS on NAS-Bench-201 further demonstrate the effectiveness and stability of AutoMCU. Real-device deployments on multiple STM32 microcontrollers validate its practical applicability to MCU-scale edge intelligence.
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.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21556unread
Beyond Single Slot: Joint Optimization for Multi-Slot Guaranteed Display Advertising
Zhaoqi Zhang, Jiaming Deng, Miao Xie, Linyou Cai, Qianlong Xie, Xingxing Wang, Siqiang Luo, Gao Cong · 2026-05-23
arXiv:2605. 21556v1 Announce Type: new Abstract: Guaranteed display advertising is crucial for platform monetization, yet existing methods often operate under a single-slot assumption, limiting their ability to optimize allocation across multi-slot page views.
Read next because Beyond Single Slot: Joint Optimization for Multi-Slot Guaranteed Display Advertising overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, fill, under, line, rate, control, test. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21556v1 Announce Type: new Abstract: Guaranteed display advertising is crucial for platform monetization, yet existing methods often operate under a single-slot assumption, limiting their ability to optimize allocation across multi-slot page views. In this paper, we propose a novel joint optimization framework for multi-slot GD allocation, addressing key challenges such as slot-level redundancy, contract imbalance, and exposure concentration. Our approach formulates the allocation as an offline bipartite matching problem with a contract roulette mechanism for slot exclusivity and Page View constraints for impression control, and incorporates a scalable allocation optimization algorithm for efficient large-scale deployment. Extensive online tests on the Meituan advertising platform demonstrate that our method significantly improves merchant ROI, platform revenue efficiency, and contract fulfillment robustness. Specifically, online A/B tests show a 28.99% increase in Average Revenue Per User under 70% traffic, and DID analysis further indicates improved contract stability, demonstrating the strong applicability and effectiveness of our framework in real-world advertising deployments.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21552unread
Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
Jinzong Dong, Zhaohui Jiang, Bo Yang · 2026-05-23
arXiv:2605. 21552v1 Announce Type: new Abstract: Confidence calibration for classification models is vital in safety-critical decision-making scenarios and has received extensive attention.
Read next because Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift 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, alignment, rate, test, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21552v1 Announce Type: new Abstract: Confidence calibration for classification models is vital in safety-critical decision-making scenarios and has received extensive attention. General confidence calibration methods assume training and test data are independent and identically distributed, limiting their effectiveness under covariate shifts. Previous calibration methods under covariate shift struggle with class-wise or canonical calibrations and often rely on unstable importance weighting when density ratios are large or unbounded. Given the above limitations, this paper rethinks confidence calibration under covariate shifts. First, we derive a necessary and sufficient condition for confidence calibration under covariate shifts, named Expectation consistency condition, which reveals covariate shifts do not necessarily lead to uncalibrated confidence and provides a weaker condition for confidence calibration than global covariate distribution alignment. Then, utilizing Expectation consistency condition, this paper proposes an unsupervised domain adaptation loss to calibrate confidence of the target domain, named Expectation consistency loss (ECL), which is compatible with canonical calibration, class-wise calibration, and top-label calibration. Third, we prove that computing ECL loss has the same sample complexity as Expected Calibration Error (ECE) and provide a theoretically grounded mini-batch trainable scheme for ECL loss. Finally, we validate the effectiveness of our method on both simulated and real-world covariate shift datasets.
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.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21550unread
PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecasting
Wangzhi Yu, Peng Zhu, Qing Zhao, Yiwen Jiang, Dawei Cheng · 2026-05-23
arXiv:2605. 21550v1 Announce Type: new Abstract: Electricity load peak forecasting (ELPF), simultaneously predicting peak timing and intensity, is a prerequisite for effective grid scheduling and risk management.
Read next because PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecasting overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, alignment, eval, line, stage. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21550v1 Announce Type: new Abstract: Electricity load peak forecasting (ELPF), simultaneously predicting peak timing and intensity, is a prerequisite for effective grid scheduling and risk management. However, existing methods face three limitations. First, they adopt a two-stage predict-then-locate paradigm, which severs the link between temporal localization and intensity regression. Second, they still struggle with the multi-scale representation conflict, leading to peak misjudgment and timing misalignment. Third, the lack of explicit peak timing context during intensity regression causes intensity smoothing because predictions are dominated by global smoothing trends. To address these limitations, we propose PeakFocus, a unified framework for ELPF. (i) A Unified Peak-Aware Pipeline (UPAP) utilizes a triple hybrid loss to jointly supervise temporal localization and intensity regression, alongside a tolerance-based evaluation protocol. (ii) A Multi-Scale Mixing Peak Locator (MSM-PL) exploits coarse-grained features to mitigate peak misjudgment caused by local fluctuations, and injects them into fine-grained features via a cascade mechanism to resolve timing misalignment. (iii) A Location-Aware Decoder (LAD) injects peak timing context into the intensity regression process, providing explicit guidance to counteract intensity smoothing and improve peak intensity estimation. Extensive experiments on the public Electricity (ELC) dataset and our industrial-scale World Large-scale Electricity Load (WLEL) dataset show that PeakFocus outperforms baselines in both timing precision and intensity estimation.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses limitation, limitations, evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21544unread
Tabular foundation models for robust calibration of near-infrared chemical sensing data
Robin Reiter, Denis Cornet, Fabien Michel, Lauriane Rouan, Gregory Beurier · 2026-05-23
arXiv:2605. 21544v1 Announce Type: new Abstract: Near-infrared spectroscopy is increasingly used as a rapid, non-destructive chemical sensing technology for the analysis of food, pharmaceutical, biological, and environmental samples.
Read next because Tabular foundation models for robust calibration of near-infrared chemical sensing data overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rect, eval, line, rate, compare, test. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21544v1 Announce Type: new Abstract: Near-infrared spectroscopy is increasingly used as a rapid, non-destructive chemical sensing technology for the analysis of food, pharmaceutical, biological, and environmental samples. However, the practical deployment of NIR sensors still depends on calibration models able to handle high-dimensional, collinear spectra, limited sample sizes, preprocessing dependence, spectral outliers, and extrapolation beyond the calibration domain. Here, we evaluate whether tabular foundation models can provide a new calibration strategy for NIR chemical sensing. We benchmark TabPFN on 66 NIR datasets covering 54 regression and 12 classification tasks, and compare direct inference on raw spectra with preprocessing-optimized inference against PLS/PLS-DA, Ridge, Catboost, and one-dimensional convolutional neural networks. The study uses a unified validation framework in which preprocessing and model selection are performed exclusively on calibration data before external test evaluation. In regression, preprocessing-optimized TabPFN achieves the best overall average rank and significantly outperforms PLS, CatBoost, TabPFN on raw spectra, and CNN-1D, while remaining statistically comparable to Ridge. In classification, TabPFN applied directly to raw spectra provides the best average rank, with performance close to the optimized variant. Robustness analyses show that TabPFN provides strong average predictive performance but that its advantage decreases on spectral outliers and extrapolated samples, where classical chemometric models remain competitive. These results suggest that tabular foundation models can complement established chemometric workflows for NIR chemical sensing, especially in small- to medium-sized calibration settings, while highlighting the need for spectroscopy-specific priors and uncertainty-aware deployment strategies.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness, evaluation, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21543unread
Provable Joint Decontamination for Benchmarking Multiple Large Language Models
Zhenlong Liu, Hao Zeng, Hongxin Wei · 2026-05-23
arXiv:2605. 21543v1 Announce Type: new Abstract: Benchmark data contamination has become a central challenge in LLM evaluation: when evaluation examples appear in the training data of one or more audited models, reported performance can be inflated and cross-model comparisons become unreliable.
Read next because Provable Joint Decontamination for Benchmarking Multiple Large Language Models overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, control, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21543v1 Announce Type: new Abstract: Benchmark data contamination has become a central challenge in LLM evaluation: when evaluation examples appear in the training data of one or more audited models, reported performance can be inflated and cross-model comparisons become unreliable. A broad line of training-data detection work designs scores to quantify how strongly a model memorizes a given data point, but these score-based methods lack theoretical guarantees. Recent conformal approaches provide provable false-identification control for a single model; however, applying them separately to each model can produce model-specific benchmarks, undermining fair comparison across models. In this work, we formalize multi-model benchmark decontamination as a joint selection problem and propose Joint Envelope Conformal Selection (JECS), a conformal procedure that enables global contamination rate (GCR) control under stated assumptions. Specifically, JECS computes per-model conformal p-values, aggregates them by the per-item maximum, and reconstructs a conservative envelope of the max-p null distribution from right-tail observations above a data-driven threshold. By applying the adaptive Benjamini-Hochberg (BH) procedure to the envelope-rescaled values, we select a benchmark with provable GCR control. Extensive experiments across various models and benchmarks demonstrate that JECS achieves higher power than the max-p baseline while consistently maintaining the target GCR control.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21542unread
Discovering Entity-Conditioned Lag Heterogeneity: A Lag-Gated Neural Audit Framework for Panel Time Series
Andi Xu · 2026-05-23
arXiv:2605. 21542v1 Announce Type: new Abstract: Country-level temporal panels are widely used in empirical analysis.
Read next because Discovering Entity-Conditioned Lag Heterogeneity: A Lag-Gated Neural Audit Framework for Panel Time Series overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, rate, test, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21542v1 Announce Type: new Abstract: Country-level temporal panels are widely used in empirical analysis. Researchers often need to audit how different entities respond to historical signals over different time horizons. Current approaches typically do not provide directly auditable entity-specific lag summaries. We formulate entity-conditioned heterogeneous lag discovery as a temporal panel mining task and propose AC-GATE, an Adaptive-Conditioning Encoder with a Scale-Invariant Lag Gate. It instantiates conditional Moderated Distributed Lag by using observable entity-level proxies to condition lag-weight distributions over historical observations, thereby making effective lags structural outputs of the model rather than post-hoc explanations. The evaluation is based on a layered audit protocol that separates predictive calibration from lag discovery. A synthetic panel with known ground-truth lags is used for mechanism recovery testing, and two real-world country-level panels are used for external audit and stress testing. The results show that AC-GATE can recover heterogeneous lag structure in synthetic data, and generates non-degenerate, externally structured effective lags in real data.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses evaluation.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21528unread
A Reproducible Log-Driven AutoML Framework for Interpretable Pipeline Optimization in Healthcare Risk Prediction
Rui Huang, Lican Huang · 2026-05-23
arXiv:2605. 21528v1 Announce Type: new Abstract: Accurate and reproducible disease risk prediction remains challenging due to heterogeneous features, limited samples, and severe class imbalance.
Read next because A Reproducible Log-Driven AutoML Framework for Interpretable Pipeline Optimization in Healthcare Risk Prediction overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, stroke, class, line, rate, full, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21528v1 Announce Type: new Abstract: Accurate and reproducible disease risk prediction remains challenging due to heterogeneous features, limited samples, and severe class imbalance. This study introduces yvsoucom-iterkit, a deterministic and log-driven automated machine learning framework that formulates pipeline optimization as a fully reproducible, configuration-level system. Each pipeline is encoded as a traceable log entity, enabling analysis of component attribution, interactions, similarity, and cross-seed robustness. Experiments on the Pima Indians Diabetes and Stroke datasets across more than 18,000 pipeline configurations reveal a structured and partially redundant search space, where performance is governed by a small subset of interacting components. Random Forest importance analysis identifies augmentation (0.454), model choice (0.198), and imbalance handling (0.101) as key drivers on Pima, while imbalance handling dominates Stroke (0.406). Component similarity analysis shows strong redundancy, with feature selection variants (biMax-biMean) exhibiting low RMS distance (0.0252), mixup closely matching no augmentation (0.0279), and TomekLinks aligning with no imbalance handling (0.0325), whereas Gaussian noise shows greater divergence from no augmentation (0.10). The framework achieves strong and stable performance using ensemble models (Weighted-F1 0.89, Macro-F1 0.88 on Pima; Weighted-F1 0.94 on Stroke), while Macro-F1 remains lower on Stroke (0.67) due to class imbalance. Cross-seed analysis reveals a performance-robustness trade-off, with ensembles showing lower variability (0.023-0.026) than SVM. These results indicate that effective AutoML optimization can focus on a reduced set of high-impact components.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21516unread
Harnesses for Inference-Time Alignment over Execution Trajectories
Boyuan Wang, Bochao Li, Minghan Wang, Yuxin Tao, Fang Kong · 2026-05-23
arXiv:2605. 21516v1 Announce Type: new Abstract: Harness engineering has emerged as an important inference-time technique for large language model (LLM) agents, aiming to improve long-term performance through task decomposition and guided execution.
Read next because Harnesses for Inference-Time Alignment over Execution Trajectories 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, rate, control, full, position, language, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21516v1 Announce Type: new Abstract: Harness engineering has emerged as an important inference-time technique for large language model (LLM) agents, aiming to improve long-term performance through task decomposition and guided execution. However, more elaborate harnesses are not uniformly better: increasing decomposition or guidance can sometimes improve execution, but can also reduce final task success. We study harness design through the lens of inference-time trajectory alignment. This perspective separates harness into two mechanisms: task decomposition, which structures a task into sub-goals, and guided execution, which reshapes local action distributions during execution. This decomposition allows us to quantify how workflow granularity, retry budgets, and guidance-induced action reweighting shape the performance limits of harness design. It further reveals concrete failure modes, including over-decomposition, over-pruning, and hallucinated execution. We validate these predictions through controlled synthetic experiments and real terminal agent benchmarks. Inspired by the theory, we further show that effective harnesses can be partial: specifying only the initial steps and leaving the remaining execution to agent can achieve higher pass rate than fully structured workflows.
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 failure, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21496unread
HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine
Brandon Dent · 2026-05-23
arXiv:2605. 21496v1 Announce Type: new Abstract: Frontier language models are being deployed into clinical workflows faster than the infrastructure to evaluate them safely.
Read next because HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, under, eval, rate, language, model, never. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21496v1 Announce Type: new Abstract: Frontier language models are being deployed into clinical workflows faster than the infrastructure to evaluate them safely. Static medical-QA benchmarks miss the failure modes that matter in emergency medicine: trajectory-level safety collapse, tool misuse, and capitulation under sustained clinical pressure. We present HealthCraft, the first public reinforcement-learning environment that rewards trajectory-level safety under realistic emergency-medicine conditions, adapted from Corecraft. It is built on a FHIR R4 world state with 14 entity types and 3,987 seed entities, exposes 24 MCP tools, and defines a dual-layer rubric that zeroes reward whenever any safety-critical criterion is violated. We release 195 tasks across six categories, graded against 2,255 binary criteria (515 safety-critical); a post-hoc 10-task negative-class slate extends this to 205 tasks and 2,337 criteria. V8 results on two frontier models show Claude Opus 4.6 at Pass@1 24.8% [21.5-28.4] and GPT-5.4 at 12.6% [10.2-15.6], with safety-failure rates of 27.5% and 34.0%. On multi-step workflows - the closest proxy to real emergency care - performance collapses to near zero (Claude 1.0%, GPT-5.4 0.0%) despite partial competence on individual steps. Six infrastructure bugs fixed between pilots v2 and v8 re-ordered which model "looks stronger," evidence that infrastructure fidelity is part of the measurement. A deterministic LLM-judge overlay bounds evaluator noise, and a 60-run negative-class smoke pilot shows the reward signal is not drop-in training-safe: restraint criteria pass at 0.929 prevalence, a gameability an eval harness can tolerate but a training reward cannot. We scaffold coupling to a Megatron+SGLang+GRPO loop per Corecraft Section 5.2 and leave training-reward ablations as future work. Environment, tasks, rubrics, and harness are released under Apache 2.0.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses failure, negative, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21493unread
Don't Collapse Your Features: Why CenterLoss Hurts OOD Detection and Multi-Scale Mahalanobis Wins
Rahul D Ray · 2026-05-23
arXiv:2605. 21493v1 Announce Type: new Abstract: The ability to detect out-of-distribution (OOD) inputs is fundamental to safe deployment of machine learning systems.
Read next because Don't Collapse Your Features: Why CenterLoss Hurts OOD Detection and Multi-Scale Mahalanobis Wins 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, trained. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21493v1 Announce Type: new Abstract: The ability to detect out-of-distribution (OOD) inputs is fundamental to safe deployment of machine learning systems. Yet, current methods often rely on feature representations that are optimised solely for classification accuracy, neglecting the distinct requirements of epistemic uncertainty. We introduce GOEN (Geometry-Optimised Epistemic Network), a simple pipeline that combines multi-scale features, L2 normalisation, Mahalanobis distance, and a calibration head trained with real hard OOD examples. Through systematic ablation we uncover a counter-intuitive finding: CenterLoss, a popular regulariser for feature compactness, significantly degrades OOD detection performance, reducing average OOD AUROC from 0.9483 to 0.9366 despite improving classification accuracy. The best variant, GOEN-NoCenterLoss, achieves an average OOD AUROC of 0.9483, surpassing all baselines including deep ensembles (0.8827), KNN (0.8967), and ODIN (0.8870) on CIFAR-10 benchmarks, while maintaining competitive in-distribution accuracy. Our results challenge the prevailing assumption that better classification geometry automatically leads to better epistemic uncertainty. Instead, we show that overly tight feature clusters compress inter-class margins and distort the covariance structure needed for effective OOD detection. GOEN is efficient, training in under 20 minutes on a single GPU, and provides a practical blueprint for building AI systems that reliably recognise their own limitations.
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, benchmark.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21492unread
The Attribution Impossibility: No Feature Ranking Is Faithful, Stable, and Complete Under Collinearity
Drake Caraker, Bryan Arnold, David Rhoads · 2026-05-23
arXiv:2605. 21492v1 Announce Type: new Abstract: No feature ranking can be simultaneously faithful, stable, and complete when features are collinear.
Read next because The Attribution Impossibility: No Feature Ranking Is Faithful, Stable, and Complete Under Collinearity 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, line, does, screen, test, model. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21492v1 Announce Type: new Abstract: No feature ranking can be simultaneously faithful, stable, and complete when features are collinear. For collinear pairs, ranking reduces to a coin flip. We prove this impossibility, quantify it for four model classes, resolve it via ensemble averaging (DASH), and machine-verify it with 305 Lean 4 theorems. We characterize the complete attribution design space: exactly two families of methods exist -- faithful-complete methods (unstable, with rankings that flip up to 50% of the time) and ensemble methods like DASH (stable, reporting ties for symmetric features) -- and no method lies outside this dichotomy. The impossibility is quantitative: the attribution ratio diverges as 1/(1-rho^2) for gradient boosting, is infinite for Lasso, and converges for random forests. DASH (Diversified Aggregation of SHAP) is provably Pareto-optimal among unbiased aggregations, achieving the Cramer-Rao variance bound with a tight ensemble size formula. In a survey of 77 public datasets, 68% exhibit attribution instability. Switching to conditional SHAP does not escape the impossibility when features have equal causal effects. The framework includes practical diagnostics -- a Z-test workflow and single-model screening tool -- and has direct consequences for fairness auditing: SHAP-based proxy discrimination audits are provably unreliable under collinearity. The design space theorem, diagnostics, and impossibility are mechanically verified in Lean 4 (305 theorems from 16 axioms, 0 sorry) -- to our knowledge, the first formally verified impossibility in explainable AI.
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 bias.
- score 100arxiv cs.LG (Machine Learning)arxiv:2605.21491unread
Teaching Language Models to Forecast Research Success Through Comparative Idea Evaluation
Srujan P Mule, Aniketh Garikaparthi, Manasi Patwardhan · 2026-05-23
arXiv:2605. 21491v1 Announce Type: new Abstract: As language models accelerate scientific research by automating hypothesis generation and implementation, a new bottleneck emerges: evaluating and filtering hundreds of AI-generated ideas without exhaustive experimentation.
Read next because Teaching Language Models to Forecast Research Success Through Comparative Idea Evaluation overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, 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, eval, rate, implement, without, candidate, test, language. Source: arxiv cs.LG (Machine Learning).
arXiv:2605.21491v1 Announce Type: new Abstract: As language models accelerate scientific research by automating hypothesis generation and implementation, a new bottleneck emerges: evaluating and filtering hundreds of AI-generated ideas without exhaustive experimentation. We ask whether LMs can learn to forecast the empirical success of research ideas before any experiments are run. We study comparative empirical forecasting: given a benchmark-specific research goal and two candidate ideas, predict which will achieve better benchmark performance. We construct a dataset of 11,488 idea pairs grounded in objective outcomes from PapersWithCode. While off-the-shelf 8B-parameter models struggle (30% acc.), SFT dramatically boosts performance to 77.1%, outperforming GPT-5 (61.1%). By framing evaluation as a reasoning task via Reinforcement Learning with Verifiable Rewards (RLVR), we train models to discover latent reasoning paths, achieving 71.35% acc. with interpretable justifications. Through additional ablations and out-of-distribution tests, we show robustness to surface-level heuristics and transfer to both a cross-domain time-split test set and an independently constructed test set. Our results demonstrate that compute-efficient small language models can serve as effective, objective verifiers, offering a scalable path for autonomous scientific discovery.
Potential threat/caveat for clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)": this item discusses robustness, evaluation, benchmark.
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.
Several of my clean results already carry MODERATE or LOW confidence flags and rest on single-seed runs or limited controls (e.g., the LoRA persona marker-leakage result, the language-mismatch SFT result); this document's proposed verification framework directly speaks to how I should interpret and qualify those findings before treating them as settled.
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.