Preventing Catastrophic Overfitting in Fast Adversarial Training: A Bi-level Optimization Perspective

Documenti analoghi

• Token-Level Constraint Boundary Search for Jailbreaking Text-to-Image Models
  di: Liu, Jiangtao, et al.
  Pubblicazione: (2025)
• Benign, Tempered, or Catastrophic: A Taxonomy of Overfitting
  di: Mallinar, Neil, et al.
  Pubblicazione: (2022)
• Implicit Jailbreak Attacks via Cross-Modal Information Concealment on Vision-Language Models
  di: Wang, Zhaoxin, et al.
  Pubblicazione: (2025)
• Enhancing the Effectiveness and Durability of Backdoor Attacks in Federated Learning through Maximizing Task Distinction
  di: Wang, Zhaoxin, et al.
  Pubblicazione: (2025)
• Improving Fast Adversarial Training Paradigm: An Example Taxonomy Perspective
  di: Gui, Jie, et al.
  Pubblicazione: (2024)