MAR-GRPO: Stabilized GRPO for AR-diffusion Hybrid Image Generation

• URL: http://arxiv.org/abs/2604.06966v1
• Date: Wed, 08 Apr 2026 11:30:35 GMT
• Title: MAR-GRPO: Stabilized GRPO for AR-diffusion Hybrid Image Generation
• Abstract summary: Reinforcement learning (RL) has been successfully applied to autoregressive (AR) and diffusion models.<n> extending RL to hybrid AR-diffusion frameworks remains challenging due to interleaved inference and noisy log-probability estimation.<n>In this work, we study masked autoregressive models (MAR) and show that the diffusion head plays a critical role in training dynamics.
• Score: 24.618644100413018
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning (RL) has been successfully applied to autoregressive (AR) and diffusion models. However, extending RL to hybrid AR-diffusion frameworks remains challenging due to interleaved inference and noisy log-probability estimation. In this work, we study masked autoregressive models (MAR) and show that the diffusion head plays a critical role in training dynamics, often introducing noisy gradients that lead to instability and early performance saturation. To address this issue, we propose a stabilized RL framework for MAR. We introduce multi-trajectory expectation (MTE), which estimates the optimization direction by averaging over multiple diffusion trajectories, thereby reducing diffusion-induced gradient noise. To avoid over-smoothing, we further estimate token-wise uncertainty from multiple trajectories and apply multi-trajectory optimization only to the top-k% uncertain tokens. In addition, we introduce a consistency-aware token selection strategy that filters out AR tokens that are less aligned with the final generated content. Extensive experiments across multiple benchmarks demonstrate that our method consistently improves visual quality, training stability, and spatial structure understanding over baseline GRPO and pre-RL models. Code is available at: https://github.com/AMAP-ML/mar-grpo.

Related papers

• DiffusionOPD: A Unified Perspective of On-Policy Distillation in Diffusion Models [55.01951088768769]
  DiffusionOPD is a new multi-task training paradigm for diffusion models based on Online Policy Distillation (OPD)<n>We show that DiffusionOPD consistently surpasses both multi-reward RL and cascade RL baselines in training efficiency and final performance.
  arXiv  Detail & Related papers  (2026-05-14T16:49:09Z)
• On the Trainability of Masked Diffusion Language Models via Blockwise Locality [58.92209096047332]
  Masked diffusion language models (MDMs) have recently emerged as a promising alternative to standard autoregressive large language models (AR-LLMs)<n>We study blockwise MDMs and compare them with AR-LLMs on three controlled tasks that stress different aspects of structured generation.<n>We find that standard random-masking MDMs fail to reliably learn linear regression, exhibit high variance training dynamics on graph path-finding, while outperforming AR-LLMs on Sudoku.
  arXiv  Detail & Related papers  (2026-04-27T17:44:26Z)
• Advantage-Guided Diffusion for Model-Based Reinforcement Learning [38.18017161791996]
  Advantage-Guided Diffusion for MBRL steers the reverse diffusion process using the agent's advantage estimates.<n>We show that the trajectories generated from AGD-MBRL follow an improved policy compared to an unguided diffusion model.
  arXiv  Detail & Related papers  (2026-04-10T06:53:25Z)
• Data-regularized Reinforcement Learning for Diffusion Models at Scale [99.01056178660538]
  We introduce Data-regularized Diffusion Reinforcement Learning ( DDRL), a novel framework that uses the forward KL divergence to anchor the policy to an off-policy data distribution.<n>With over a million GPU hours of experiments and ten thousand double-blind evaluations, we demonstrate that DDRL significantly improves rewards while alleviating the reward hacking seen in RLs.
  arXiv  Detail & Related papers  (2025-12-03T23:45:07Z)
• Enhancing Diffusion-based Restoration Models via Difficulty-Adaptive Reinforcement Learning with IQA Reward [93.04811239892852]
  Reinforcement Learning (RL) has recently been incorporated into diffusion models.<n>In this paper, we investigate how to effectively integrate RL into diffusion-based restoration models.
  arXiv  Detail & Related papers  (2025-11-03T14:57:57Z)
• SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation [62.14510717860079]
  We propose a Synergistic Diffusion-Autoregression paradigm that unifies the training efficiency of autoregressive models with the parallel inference capability of diffusion.<n>SDAR performs a lightweight paradigm conversion that transforms a well-trained autoregressive (AR) model into a blockwise diffusion model through brief, data-efficient adaptation.<n>Building on this insight, SDAR achieves efficient AR-to-diffusion conversion with minimal cost, preserving AR-level performance while enabling parallel generation.
  arXiv  Detail & Related papers  (2025-10-07T17:29:28Z)
• Consolidating Reinforcement Learning for Multimodal Discrete Diffusion Models [40.82263997290613]
  We introduce MaskGRPO, the first viable approach to enable scalable multimodal reinforcement learning in discrete diffusion.<n>MaskGRPO brings more stable and efficient updates, leading to stronger reasoning performance and better generation quality.
  arXiv  Detail & Related papers  (2025-10-03T10:36:24Z)
• EADReg: Probabilistic Correspondence Generation with Efficient Autoregressive Diffusion Model for Outdoor Point Cloud Registration [22.084204833166982]
  We propose a novel framework named EADReg for efficient and robust registration of LiDAR point clouds. We employ a Bi-directional Gaussian Mixture Model (BGMM) to reject outlier points and obtain purified point cloud pairs. In the fine stage, we treat diffusion-based PCR as an autoregressive process to generate robust point correspondences.
  arXiv  Detail & Related papers  (2024-11-22T13:03:54Z)
• Avoiding mode collapse in diffusion models fine-tuned with reinforcement learning [0.0]
  Fine-tuning foundation models via reinforcement learning (RL) has proven promising for aligning to downstream objectives. We exploit the hierarchical nature of diffusion models (DMs) and train them dynamically at each epoch with a tailored RL method. We show that models trained with HRF achieve better preservation of diversity in downstream tasks, thus enhancing the fine-tuning robustness and at uncompromising mean rewards.
  arXiv  Detail & Related papers  (2024-10-10T19:06:23Z)
• Understanding Reinforcement Learning-Based Fine-Tuning of Diffusion Models: A Tutorial and Review [63.31328039424469]
  This tutorial provides a comprehensive survey of methods for fine-tuning diffusion models to optimize downstream reward functions. We explain the application of various RL algorithms, including PPO, differentiable optimization, reward-weighted MLE, value-weighted sampling, and path consistency learning.
  arXiv  Detail & Related papers  (2024-07-18T17:35:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.