Amplification Effects in Test-Time Reinforcement Learning: Safety and Reasoning Vulnerabilities

Documenti analoghi

• Forget to Flourish: Leveraging Machine-Unlearning on Pretrained Language Models for Privacy Leakage
  di: Rashid, Md Rafi Ur, et al.
  Pubblicazione: (2024)
• Directional Embedding Smoothing for Robust Vision Language Models
  di: Wang, Ye, et al.
  Pubblicazione: (2026)
• Efficient Differentially Private Fine-Tuning of Diffusion Models
  di: Liu, Jing, et al.
  Pubblicazione: (2024)
• Smoothed Embeddings for Robust Language Models
  di: Hase, Ryo, et al.
  Pubblicazione: (2025)
• Variational Randomized Smoothing for Sample-Wise Adversarial Robustness
  di: Hase, Ryo, et al.
  Pubblicazione: (2024)