Current Focus
- Security and trustworthiness of LLM-based agents
- Adversarial attacks, context and memory manipulation
- Behavioral risks in multi-step decision-making
- Privacy leakage in interactive agent systems
About Me
Research Interests
Research Foundations
- Backdoor attacks and temporal attack surfaces
- Federated learning security and privacy protection
- Model debiasing, alignment, and interpretability
- Secure communication protocols for IoT systems
News
Recent Updates
Publications
Selected Work
Education
Education
Service