EPISODE · Jan 7, 2026 · 22 MIN
VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation
from Daily Paper Cast · host Jingwen Liang, Gengyu Wang
🤗 Upvotes: 29 | cs.CV, cs.LG Authors: Shikun Sun, Liao Qu, Huichao Zhang, Yiheng Liu, Yangyang Song, Xian Li, Xu Wang, Yi Jiang, Daniel K. Du, Xinglong Wu, Jia Jia Title: VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation Arxiv: http://arxiv.org/abs/2601.02256v1 Abstract: Visual generation is dominated by three paradigms: AutoRegressive (AR), diffusion, and Visual AutoRegressive (VAR) models. Unlike AR and diffusion, VARs operate on heterogeneous input structures across their generation steps, which creates severe asynchronous policy conflicts. This issue becomes particularly acute in reinforcement learning (RL) scenarios, leading to unstable training and suboptimal alignment. To resolve this, we propose a novel framework to enhance Group Relative Policy Optimization (GRPO) by explicitly managing these conflicts. Our method integrates three synergistic components: 1) a stabilizing intermediate reward to guide early-stage generation; 2) a dynamic time-step reweighting scheme for precise credit assignment; and 3) a novel mask propagation algorithm, derived from principles of Reward Feedback Learning (ReFL), designed to isolate optimization effects both spatially and temporally. Our approach demonstrates significant improvements in sample quality and objective alignment over the vanilla GRPO baseline, enabling robust and effective optimization for VAR models.
What this episode covers
🤗 Upvotes: 29 | cs.CV, cs.LG Authors: Shikun Sun, Liao Qu, Huichao Zhang, Yiheng Liu, Yangyang Song, Xian Li, Xu Wang, Yi Jiang, Daniel K. Du, Xinglong Wu, Jia Jia Title: VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation Arxiv: http://arxiv.org/abs/2601.02256v1 Abstract: Visual generation is dominated by three paradigms: AutoRegressive (AR), diffusion, and Visual AutoRegressive (VAR) models. Unlike AR and diffusion, VARs operate on heterogeneous input structures across their generation steps, which creates severe asynchronous policy conflicts. This issue becomes particularly acute in reinforcement learning (RL) scenarios, leading to unstable training and suboptimal alignment. To resolve this, we propose a novel framework to enhance Group Relative Policy Optimization (GRPO) by explicitly managing these conflicts. Our method integrates three synergistic components: 1) a stabilizing intermediate reward to guide early-stage generation; 2) a dynamic time-step reweighting scheme for precise credit assignment; and 3) a novel mask propagation algorithm, derived from principles of Reward Feedback Learning (ReFL), designed to isolate optimization effects both spatially and temporally. Our approach demonstrates significant improvements in sample quality and objective alignment over the vanilla GRPO baseline, enabling robust and effective optimization for VAR models.
NOW PLAYING
VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation
No transcript for this episode yet
Similar Episodes
No similar episodes found.