REVELIO uncovers interpretable failure modes in VLMs by searching combinatorial concept spaces with diversity-aware beam search and Gaussian-process Thompson sampling, revealing vulnerabilities in autonomous driving and indoor robotics.
Safety alignment for vision language models, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Revealing Interpretable Failure Modes of VLMs
REVELIO uncovers interpretable failure modes in VLMs by searching combinatorial concept spaces with diversity-aware beam search and Gaussian-process Thompson sampling, revealing vulnerabilities in autonomous driving and indoor robotics.