← authors · overview

• Defending Pre-trained Language Models As Few-shot Learners Against Backdoor Attacks
  2023 · 12 citations
• G$^2$uardFL: Safeguarding Federated Learning Against Backdoor Attacks through Attributed Client Graph Clustering
  2023 · 8 citations
• G$^2$uardFL: Safeguarding Federated Learning Against Backdoor Attacks through Attributed Client Graph Clustering
  2023 · 8 citations
• Lilith: Backdoor Generalization under Training-Inference Trigger Shift
  2026
• NeuroBreak: Unveil Internal Jailbreak Mechanisms in Large Language Models
  2025
• Automatic Red Teaming LLM-based Agents with Model Context Protocol Tools
  2025