Yige Li

Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models

Propaganda AI: An Analysis of Semantic Divergence in Large Language Models

Where Did It Go Wrong? Attributing Undesirable LLM Behaviors via Representation Gradient Tracing

AgenticPA: Toward Automated and Large-Scale Prompt Attacks on LLMs

AutoBackdoor: Automating Backdoor Attacks via LLM Agents

Memory Injection Attacks on LLM Agents via Query-Only Interaction

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization

Detecting Backdoor Samples in Contrastive Language Image Pretraining

BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks

Adversarial Suffixes May Be Features Too!

AnyAttack: Self-supervised Generation of Targeted Adversarial Attacks for Vision-Language Models

TUAP: Targeted Universal Adversarial Perturbations for CLIP

Do Influence Functions Work on Large Language Models?