Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning

Documenti analoghi

• Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning
  di: Fan, Chongyu, et al.
  Pubblicazione: (2024)
• Transformers as Decision Makers: Provable In-Context Reinforcement Learning via Supervised Pretraining
  di: Lin, Licong, et al.
  Pubblicazione: (2023)
• Stable Preference Optimization: A Bilevel Approach to Catastrophic Preference Shift
  di: Jian, Chengtao, et al.
  Pubblicazione: (2025)
• Bridging the Gap Between Preference Alignment and Machine Unlearning
  di: Feng, Xiaohua, et al.
  Pubblicazione: (2025)
• ZeroUnlearn: Few-Shot Knowledge Unlearning in Large Language Models
  di: Lin, Yujie, et al.
  Pubblicazione: (2026)