Poster · Sydney
Missing Old Logits in Asynchronous Agentic RL: Semantic Mismatch and Repair Methods for Off-Policy Correction
- 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
- 钟 关, Yongjian Guo, Haoran Sun, Wen Huang, shuai di, Junwu Xiong, Likang Wu, Hongke Zhao
- Organisations
- Tianjin University, Tsinghua University, Peking University, JD.com, JD Technology
- QRS classification
- Reinforcement learning, Agentic AI