Si Yuan Liang

FERD: Fairness-Enhanced Data-Free Adversarial Robustness Distillation

E$^2$AT: Multimodal Jailbreak Defense via Dynamic Joint Optimization

RoboView-Bias: Benchmarking Visual Bias in Embodied Agents for Robotic Manipulation

Where MLLMs Attend and What They Rely On: Explaining Autoregressive Token Generation