arXiv cs.CV
· Papers
Policy-based Tuning of Autoregressive Image Models with Instance- and Distribution-Level Rewards
arXiv:2603.23086v2 Announce Type: replace-cross Abstract: Autoregressive (AR) models are highly effective for image generation, yet their standard maximum-likelihood estimation training lacks direct optimization for sample quality and diversity. While reinforcement learning (RL) has been used to align diffusion models,