Poster · Sydney
Efficient Off-Policy RL for Video Generation via Forward-Consistent Reward Matching
- 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
- Hongzheng Yang, Mengyang LIU, Haoxuan Wu, Kun Li, Yuzhi Zhao, Wei Liu
- Organisations
- The Chinese University of Hong Kong, Video Rebirth, City University of Hong Kong, Huawei, Tencent
- QRS classification
- Reinforcement learning, Generative models, Computer vision