Fetching the paper…

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift · Around