Poster · Sydney
Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space
- When
- Tue 8 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 1
- Authors
- Hengli Li, Chenxi Li, Tong Wu, Xuekai Zhu, Yuxuan Wang, Zhaoxin Yu, Eric Hanchen Jiang, Song-Chun Zhu, Zixia Jia, Ying Nian Wu, Zilong Zheng
- Organisations
- Peking University, Tsinghua University, Beijing Institute for General Artificial Intelligence, Shanghai Jiao Tong University, Alibaba, Chinese Academy of Sciences, University of California, Los Angeles, BigAI
- QRS classification
- Reinforcement learning, Language models