Peizhi Niu

GUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models

MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools

DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform

Do LLMs Really Forget? Evaluating Unlearning with Knowledge Correlation and Confidence Awareness

Improving Defense Mechanisms for Subgraph-Structure Membership Inference Attacks