類似資料

• SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
  著者:: He, Xuanli, 等
  出版事項: (2024)
• TuBA: Cross-Lingual Transferability of Backdoor Attacks in LLMs with Instruction Tuning
  著者:: He, Xuanli, 等
  出版事項: (2024)
• Defending against Backdoor Attacks via Module Switching
  著者:: Li, Weijun, 等
  出版事項: (2025)
• Cut the Deadwood Out: Backdoor Purification via Guided Module Substitution
  著者:: Tong, Yao, 等
  出版事項: (2024)
• Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge
  著者:: Arora, Ansh, 等
  出版事項: (2024)