Jiaming Zhang

Adversarial Prompt Tuning for Vision-Language Models

Low-mid adversarial perturbation against unauthorized face recognition system

Introducing Foundation Models as Surrogate Models: Advancing Towards More Practical Adversarial Attacks

ImageNet pre-training also transfers non-robustness

Attention, Please! Adversarial Defense via Activation Rectification and Preservation

Unlearnable clusters: Towards label-agnostic unlearnable examples

Benign Adversarial Attack: Tricking Models for Goodness

Towards adversarial attack on vision-language pre-training models