It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank
Frozen vision-language models increasingly provide safety signals for reinforcement learning.
ProofPaper ↗
Key points
- Their use assumes that similarity to language describing danger indicates the hazard itself.
- VLM-based methods have reported gains in driving and safe-RL benchmarks by converting image-text similarity into rewards, costs, or confidence weights.
- To address this gap, we present a controlled evaluation of a frozen CLIP prompt-margin safety score.
- We apply the score to trajectories generated by policies that never receive it, match pre-contact observations to contact-free observations with comparable hazard geometry, and vary the captions, encoder, and camera view.
Sources (1)
- [1]It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption BankarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 06:15 AM
Frozen vision-language models increasingly provide safety signals for reinforcement learning.
Their use assumes that similarity to language describing danger indicates the hazard itself.
Extractive summary: sentences quoted from the sources.