Jizhong Han

Aligning at the Source: Steering Corrective to the Origins of Harmfulness in LLMs

Mitigating Object Hallucinations in Large Vision-Language Models via multi-scale visual integration

Resolving the Security-Auditability Dilemma with Auditable Latent Chain-of-Thought

Modulating LLM Behavior via Context-Specific Activation Steering

Resolution Attack: Exploiting Image Compression to Deceive Deep Neural Networks

OMS: One More Step Noise Searching to Enhance Membership Inference Attacks for Diffusion Models