Paper

Forecasting Future Behavior as a Learning Task

Unreadunread

AI summary

arXiv:2606. 11445v1 Announce Type: new Abstract: Trust in an AI system is often anchored by explanations of how it works, which one then uses to forecast its behavior on new inputs.

Reading status

Priority

About this paper

Authors: Mosh Levy, Yoav Goldberg, Asa Cooper Stickland
Released: 2026-06-11 (28d ago)
Topic: Related to current project
Relation to my work: Read next because Forecasting Future Behavior as a Learning Task 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, eval, token, rate, trained, language, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv: 2606.11445

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to Forecasting Future Behavior as a Learning Task. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.