Poster · Sydney
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information
- When
- Wed 9 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 3
- Authors
- Guobin Shen, Xiang Cheng, Chenxiao Zhao, Lei Huang, Jindong Li, Dongcheng Zhao, XingYu Li
- Organisations
- ByteDance, Xiaohongshu, Harbin Institute of Technology, Chinese Academy of Sciences
- QRS classification
- Reinforcement learning