Adversarial Vulnerability as a Consequence of On-Manifold Inseparibility

Documenti analoghi

• Effect of Ambient-Intrinsic Dimension Gap on Adversarial Vulnerability
  di: Haldar, Rajdeep, et al.
  Pubblicazione: (2024)
• f-GRPO and Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM Alignment
  di: Haldar, Rajdeep, et al.
  Pubblicazione: (2026)
• LLM Safety Alignment is Divergence Estimation in Disguise
  di: Haldar, Rajdeep, et al.
  Pubblicazione: (2025)
• Better Representations via Adversarial Training in Pre-Training: A Theoretical Perspective
  di: Xing, Yue, et al.
  Pubblicazione: (2024)
• Fair Supervised Learning with A Simple Random Sampler of Sensitive Attributes
  di: Sohn, Jinwon, et al.
  Pubblicazione: (2023)