Adversarial training at both the CLIP pre-training stage and the LLaVA instruction-tuning stage produces vision-language models with state-of-the-art robustness and near-baseline clean performance.
Image hijacking: Adversarial images can control generative models at runtime
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.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness
Adversarial training at both the CLIP pre-training stage and the LLaVA instruction-tuning stage produces vision-language models with state-of-the-art robustness and near-baseline clean performance.