The Invitation Trap: Proactive Availability Backdoor in LLMs via Conversational Induction

Documenti analoghi

• BackdoorIndicator: Leveraging OOD Data for Proactive Backdoor Detection in Federated Learning
  di: Li, Songze, et al.
  Pubblicazione: (2024)
• TrapSuffix: Proactive Defense Against Adversarial Suffixes in Jailbreaking
  di: Du, Mengyao, et al.
  Pubblicazione: (2026)
• Stealthy Backdoor Attack via Confidence-driven Sampling
  di: He, Pengfei, et al.
  Pubblicazione: (2023)
• A Practical Trigger-Free Backdoor Attack on Neural Networks
  di: Wang, Jiahao, et al.
  Pubblicazione: (2024)
• Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMs
  di: Li, Yige, et al.
  Pubblicazione: (2026)