Chaohu Liu

AdPO: Enhancing the Adversarial Robustness of Large Vision-Language Models with Preference Optimization

ChartMaster: Boosting MLLMs for Chart Analysis through Data, Perception, and Reasoning Optimization

DARE:Difficulty-Aware Dynamic Routing for Mixture of Experts

FPEdit: Robust LLM Fingerprinting through Localized Parameter Editing

Tracking the Copyright of Large Vision-Language Models through Parameter Learning Adversarial Images

AdPO: Enhancing the Adversarial Robustness of Large Vision-Language Models with Preference Optimization

Gradient Regularization-based Cross-Prompt Attacks on Vision Language Models