{"ID":2855147,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.13418","arxiv_id":"2510.13418","title":"Reinforcement Learning Meets Masked Generative Models: Mask-GRPO for Text-to-Image Generation","abstract":"Reinforcement learning (RL) has garnered increasing attention in text-to-image (T2I) generation. However, most existing RL approaches are tailored to either diffusion models or autoregressive models, overlooking an important alternative: masked generative models. In this work, we propose Mask-GRPO, the first method to incorporate Group Relative Policy Optimization (GRPO)-based RL into this overlooked paradigm. Our core insight is to redefine the transition probability, which is different from current approaches, and formulate the unmasking process as a multi-step decision-making problem. To further enhance our method, we explore several useful strategies, including removing the KL constraint, applying the reduction strategy, and filtering out low-quality samples. Using Mask-GRPO, we improve a base model, Show-o, with substantial improvements on standard T2I benchmarks and preference alignment, outperforming existing state-of-the-art approaches. The code is available on https://github.com/xingzhejun/Mask-GRPO","short_abstract":"Reinforcement learning (RL) has garnered increasing attention in text-to-image (T2I) generation. However, most existing RL approaches are tailored to either diffusion models or autoregressive models, overlooking an important alternative: masked generative models. In this work, we propose Mask-GRPO, the first method to...","url_abs":"https://arxiv.org/abs/2510.13418","url_pdf":"https://arxiv.org/pdf/2510.13418v2","authors":"[\"Yifu Luo\",\"Xinhao Hu\",\"Keyu Fan\",\"Haoyuan Sun\",\"Zeyu Chen\",\"Bo Xia\",\"Tiantian Zhang\",\"Yongzhe Chang\",\"Xueqian Wang\"]","published":"2025-10-15T11:18:12Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Reinforcement Learning\",\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":608223,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2855147,"paper_url":"https://arxiv.org/abs/2510.13418","paper_title":"Reinforcement Learning Meets Masked Generative Models: Mask-GRPO for Text-to-Image Generation","repo_url":"https://github.com/xingzhejun/Mask-GRPO","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}