Vision-language model judgments about image realism and generated captions can be shifted by imperceptible perturbations confined to specific spatial frequency bands, even under black-box access.
Semantically consistent visual representation for adversarial robustness,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Reliability of Vision-Language Models Under Adversarial Frequency-Domain Perturbations
Vision-language model judgments about image realism and generated captions can be shifted by imperceptible perturbations confined to specific spatial frequency bands, even under black-box access.