arxiv
PublishedMay 4, 2026 at 4:00 AM
—neutral
Prompt-Induced Score Variance in Zero-Shot Binary Vision-Language Safety Classification
Publisher summary· verbatim
arXiv:2605.00326v1 Announce Type: new Abstract: Single-prompt first-token probabilities from zero-shot vision-language model (VLM) safety classifiers are treated as decision scores, but we show they are unreliable under semantically equivalent prompt reformulation: even when the binary label is cons
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
The Bubble Brief
WEEKLYRead safety insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗