A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers

Documenti analoghi

• Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
  di: Nguyen, Quang H., et al.
  Pubblicazione: (2024)
• When Forgetting Triggers Backdoors: A Clean Unlearning Attack
  di: Arazzi, Marco, et al.
  Pubblicazione: (2025)
• FFCBA: Feature-based Full-target Clean-label Backdoor Attacks
  di: Yin, Yangxu, et al.
  Pubblicazione: (2025)
• Mitigating Backdoor Triggered and Targeted Data Poisoning Attacks in Voice Authentication Systems
  di: Mohammadi, Alireza, et al.
  Pubblicazione: (2025)
• Checkerboard: A Simple, Effective, Efficient and Learning-free Clean Label Backdoor Attack with Low Poisoning Budget
  di: Yang, Yi, et al.
  Pubblicazione: (2026)