Yun Shen

The paper analyzes that while multimodal large language models (MLLMs) offer superior semantic understanding for image generation, this enhanced capability significantly increases safety risks, particularly in generating unsafe content and creating harder-to-detect fake images compared to traditional diffusion models.

The paper investigates how various fine-tuning methods can be used both to intentionally misalign and subsequently realign large language models (LLMs), revealing distinct strengths for attack and defense mechanisms.

The paper introduces BadBone, a stealthy and adaptive backdoor attack that compromises a backbone model specifically to target downstream tasks utilizing prompt learning, demonstrating high attack success rates against state-of-the-art defenses.