Podobne knjige/članki

• NSmark: Null Space Based Black-box Watermarking Defense Framework for Language Models
  od: Zhao, Haodong, et al.
  Izdano: (2024)
• SynGhost: Invisible and Universal Task-agnostic Backdoor Attack via Syntactic Transfer
  od: Cheng, Pengzhou, et al.
  Izdano: (2024)
• Acquiring Clean Language Models from Backdoor Poisoned Datasets by Downscaling Frequency Space
  od: Wu, Zongru, et al.
  Izdano: (2024)
• Gracefully Filtering Backdoor Samples for Generative Large Language Models without Retraining
  od: Wu, Zongru, et al.
  Izdano: (2024)
• Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models
  od: Zhao, Tianhang, et al.
  Izdano: (2025)