FDA differentially subtracts function-word cross-attention from original attention heads to cut attack success rates by 18-90% across models and tasks while dropping performance by at most 0.6%.
Sa-attack: Improving adversar- ial transferability of vision-language pre-training models via self-augmentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Pay Less Attention to Function Words for Free Robustness of Vision-Language Models
FDA differentially subtracts function-word cross-attention from original attention heads to cut attack success rates by 18-90% across models and tasks while dropping performance by at most 0.6%.