VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models

Devs

VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models | Read Paper on Bytez