Improving Adversarial Transferability of Vision-Language Pre-training Models through Collaborative Multimodal Interaction

VLM
AAML
Re-assign community