Mechanistic Exploration of Backdoored Large Language Model Attention Patterns

Ähnliche Einträge

• Unmasking Backdoors: An Explainable Defense via Gradient-Attention Anomaly Scoring for Pre-trained Language Models
  von: Das, Anindya Sundar, et al.
  Veröffentlicht: (2025)
• Composite Backdoor Attacks Against Large Language Models
  von: Huang, Hai, et al.
  Veröffentlicht: (2023)
• Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models
  von: Xu, Jiashu, et al.
  Veröffentlicht: (2023)
• Binary Autoencoder for Mechanistic Interpretability of Large Language Models
  von: Cho, Hakaze, et al.
  Veröffentlicht: (2025)
• Mechanistic Anomaly Detection for "Quirky" Language Models
  von: Johnston, David O., et al.
  Veröffentlicht: (2025)