OP-GRPO: Efficient Off-Policy GRPO for Flow-Matching Models

• URL: http://arxiv.org/abs/2604.04142v1
• Date: Sun, 05 Apr 2026 15:00:29 GMT
• Title: OP-GRPO: Efficient Off-Policy GRPO for Flow-Matching Models
• Abstract summary: We present OP-GRPO, the first Off-Policy GRPO framework tailored for flow-matching models.<n>We actively select high-quality trajectories and adaptively incorporate them into a replay buffer for reuse in subsequent training iterations.
• Score: 14.396100082949005
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Post training via GRPO has demonstrated remarkable effectiveness in improving the generation quality of flow-matching models. However, GRPO suffers from inherently low sample efficiency due to its on-policy training paradigm. To address this limitation, we present OP-GRPO, the first Off-Policy GRPO framework tailored for flow-matching models. First, we actively select high-quality trajectories and adaptively incorporate them into a replay buffer for reuse in subsequent training iterations. Second, to mitigate the distribution shift introduced by off-policy samples, we propose a sequence-level importance sampling correction that preserves the integrity of GRPO's clipping mechanism while ensuring stable policy updates. Third, we theoretically and empirically show that late denoising steps yield ill-conditioned off-policy ratios, and mitigate this by truncating trajectories at late steps. Across image and video generation benchmarks, OP-GRPO achieves comparable or superior performance to Flow-GRPO with only 34.2% of the training steps on average, yielding substantial gains in training efficiency while maintaining generation quality.

Related papers

• AdaGRPO: A Capability-Aware Adaptive Enhancement for Flow-based GRPO [78.36537400975298]
  Group Relative Policy Optimization (GRPO) has demonstrated remarkable success in aligning text-to-image (T2I) flow models with human preferences.<n>We have identified that the learning loop of current flow-based GRPO is fundamentally decoupled from the learner's current capability.<n>We propose Adaptive GRPO (AdaGRPO), a novel capability-aware RL algorithm tailored for flow models.
  arXiv  Detail & Related papers  (2026-06-05T02:07:08Z)
• BiasGRPO: Stabilizing Bias Mitigation in High-Variance Reward Landscapes via Group-Relative Policy Optimization [23.13988703785064]
  Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge.<n>Direct Preference Optimization (DPO) is limited by the lack of exploration inherent in offline training.<n>Proximal Policy Optimization (PPO) can lead to training instability due to potentially unreliable critic estimates.
  arXiv  Detail & Related papers  (2026-06-03T12:31:42Z)
• Sample-Efficient Diffusion-based Reinforcement Learning with Critic Guidance [38.06932977050757]
  We propose CGPO, textbfCritic-textbfGuided diffusion textbfPolicy textbfOptimization.<n>CGPO steers action generation toward high-value regions defined by the critic network and uses the guided actions as regression objectives.<n>We validate the effectiveness of CGPO on 5 MuJoCo locomotion tasks, and CGPO achieves state-of-the-art performance compared with existing diffusion-based RL methods.
  arXiv  Detail & Related papers  (2026-05-28T15:07:50Z)
• V-GRPO: Online Reinforcement Learning for Denoising Generative Models Is Easier than You Think [90.69263509098948]
  We introduce Variational GRPO, a method that integrates ELBO-based surrogates with the Group Relative Policy Optimization algorithm.<n>V-GRPO achieves state-of-the-art performance in text-to-image synthesis, while delivering a $2times$ speedup over MixGRPO and a $3times$ speedup over DiffusionNFT.
  arXiv  Detail & Related papers  (2026-04-25T17:03:21Z)
• Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization [60.87651283510059]
  Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs.<n>We propose Dynamic Pruning Policy Optimization (DPPO), a framework that enables dynamic pruning while preserving unbiased gradient estimation.<n>To mitigate the data sparsity induced by pruning, we introduce Dense Prompt Packing, a window-based greedy strategy.
  arXiv  Detail & Related papers  (2026-03-04T14:48:53Z)
• iGRPO: Self-Feedback-Driven LLM Reasoning [88.83313431248473]
  Large Language Models (LLMs) have shown promise in solving complex mathematical problems, yet they still fall short of producing accurate and consistent solutions.<n>We introduce Iterative Group Relative Policy Optimization (iGRPO), a two-stage extension of GRPO that adds dynamic self-conditioning through model-generated drafts.<n>Under matched rollout budgets, iGRPO consistently outperforms GRPO across base models.
  arXiv  Detail & Related papers  (2026-02-09T18:45:11Z)
• GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping [63.33669214116784]
  GRPO-Guard is a simple yet effective enhancement to existing GRPO frameworks.<n>It restores a balanced and step-consistent importance ratio, ensuring that PPO clipping properly constrains harmful updates.<n>It substantially mitigates implicit over-optimization without relying on heavy KL regularization.
  arXiv  Detail & Related papers  (2025-10-25T14:51:17Z)
• BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping [69.74252624161652]
  We propose BAlanced Policy Optimization with Adaptive Clipping (BAPO)<n>BAPO dynamically adjusts clipping bounds to adaptively re-balance positive and negative contributions, preserve entropy, and stabilize RL optimization.<n>On AIME 2024 and AIME 2025 benchmarks, our 7B BAPO model surpasses open-source counterparts such as SkyWork-OR1-7B.
  arXiv  Detail & Related papers  (2025-10-21T12:55:04Z)
• BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models [57.304411396229035]
  We present BranchGRPO, a method that restructures the rollout process into a branching tree.<n>On HPDv2.1 image alignment, BranchGRPO improves alignment scores by up to textbf16% over DanceGRPO.<n>A hybrid variant, BranchGRPO-Mix, further accelerates training to 4.7x faster than DanceGRPO without degrading alignment.
  arXiv  Detail & Related papers  (2025-09-07T12:53:06Z)
• TempFlow-GRPO: When Timing Matters for GRPO in Flow Models [22.023027865557637]
  We introduce a principled GRPO framework that captures and exploits the temporal structure inherent in flow-based generation.<n>New innovations endow the model with temporally-aware optimization that respects the underlying generative dynamics.
  arXiv  Detail & Related papers  (2025-08-06T11:10:39Z)

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.