6 papers
SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models
Jiesong Lian, Zixiang Zhou, Ruizhe Zhong +6
Recent video diffusion models (VDMs) synthesize visually convincing clips, yet still drop entities, mis-bind attributes, and weaken the interactions specified in the prompt. Repres…
Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level
Nan Jia, Haojin Yang, Xing Ma +6
On-policy distillation (OPD) trains a student on its own trajectories with token-level teacher feedback and often outperforms off-policy distillation and standard reinforcement lea…
SoliReward: Mitigating Susceptibility to Reward Hacking and Annotation Noise in Video Generation Reward Models
Jiesong Lian, Ruizhe Zhong, Zixiang Zhou +6
Post-training alignment of video generation models with human preferences is a critical goal. Developing effective Reward Models (RMs) for this process faces significant methodolog…
Euphonium: Steering Video Flow Matching via Process Reward Gradient Guided Stochastic Dynamics
Ruizhe Zhong, Jiesong Lian, Xiaoyue Mi +4
While online Reinforcement Learning has emerged as a crucial technique for aligning flow matching models with human preferences, current approaches are hindered by inefficient expl…
Video Generation Models Are Good Latent Reward Models
Xiaoyue Mi, Wenqing Yu, Jiesong Lian +9
Reward feedback learning (ReFL) has proven effective for aligning image generation with human preferences. However, its extension to video generation faces significant challenges.…
Fusion-PSRO: Nash Policy Fusion for Policy Space Response Oracles
Jiesong Lian, Yucong Huang, Chengdong Ma +4
For solving zero-sum games involving non-transitivity, a useful approach is to maintain a policy population to approximate the Nash Equilibrium (NE). Previous studies have shown th…