My research interest focuses on designing Trustworthy Agentic AI by understanding from theory to implementation and by considering practical applications in robotics and computer security from a learning-theoretical perspective.

Trustworthy LLMs
Trustworthy LLMs

Make LLMs trustful, fair, safe, and secue.

Trustworthy Code Generation
Trustworthy Code Generation

Build functional and vulerability-free Code Generators.

Trustworthy Robotics
Trustworthy Robotics

Live with Safe Robots.

Red Teaming on Agentic AI
Red Teaming on Agentic AI

Make Agentic AI better by Red Teaming

Machine Unlearning
Machine Unlearning

Build a hammer for Trustworthy Agentic AI

Uncertainty Quantification
Uncertainty Quantification

Trustworthy decision making with reliable uncertainty.