Paper

Seeing Before Colliding: Anticipatory Safe RL with Frozen Vision-Language Models

Unreadunread

AI summary

arXiv:2606. 11266v1 Announce Type: new Abstract: The cost signal that constrained-RL algorithms optimize against is almost always reactive: the simulator emits a non-zero cost only after a collision has begun, and the Lagrange multiplier of PPO-Lagrangian grows only after the episode budget has been exceeded.

Reading status

Priority

About this paper

Authors: Samuel Tetteh, Cody Fleming
Released: 2026-06-12 (27d ago)
Topic: Related to current project
Relation to my work: Read next because Seeing Before Colliding: Anticipatory Safe RL with Frozen Vision-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: text, rect, eval, line, rate, does, trained, emit. Source: arxiv cs.LG (Machine Learning).
arXiv: 2606.11266

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to Seeing Before Colliding: Anticipatory Safe RL with Frozen Vision-Language Models. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.