← authors · overview

• Drattack: Prompt Decomposition And Reconstruction Makes Powerful LLM Jailbreakers
  2024 · 103 citations
• Sign-opt: A Query-efficient Hard-label Adversarial Attack
  2019 · 29 citations
• One Prompt Is Not Enough: Automated Construction Of A Mixture-of-expert Prompts
  2024 · 23 citations
• DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers
  2024 · 4 citations
• FedDM: Iterative Distribution Matching for Communication-Efficient Federated Learning
  2022 · 3 citations
• Revisiting Personalized Federated Learning: Robustness Against Backdoor Attacks
  2023 · 1 citations
• CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs
  2026
• PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt Sanitization
  2025
• TASO: Jailbreak LLMs via Alternative Template and Suffix Optimization
  2025
• R1-Zero's "Aha Moment" in Visual Reasoning on a 2B Non-SFT Model
  2025
• Exploring Expert Failures Improves LLM Agent Tuning
  2025
• ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
  2025
• A Review of Adversarial Attack and Defense for Classification Methods
  2021