VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models

AAMLVLMRe-assign community