TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Flow-GRPO: Training Flow Matching Models via Online RL

Flow-GRPO: Training Flow Matching Models via Online RL

Jie Liu, Gongye Liu, Jiajun Liang, Yangguang Li, Jiaheng Liu, Xintao Wang, Pengfei Wan, Di Zhang, Wanli Ouyang

2025-05-08DenoisingText-to-Image GenerationText GenerationReinforcement Learning
PaperPDFCode(official)

Abstract

We propose Flow-GRPO, the first method integrating online reinforcement learning (RL) into flow matching models. Our approach uses two key strategies: (1) an ODE-to-SDE conversion that transforms a deterministic Ordinary Differential Equation (ODE) into an equivalent Stochastic Differential Equation (SDE) that matches the original model's marginal distribution at all timesteps, enabling statistical sampling for RL exploration; and (2) a Denoising Reduction strategy that reduces training denoising steps while retaining the original inference timestep number, significantly improving sampling efficiency without performance degradation. Empirically, Flow-GRPO is effective across multiple text-to-image tasks. For complex compositions, RL-tuned SD3.5 generates nearly perfect object counts, spatial relations, and fine-grained attributes, boosting GenEval accuracy from 63% to 95%. In visual text rendering, its accuracy improves from 59% to 92%, significantly enhancing text generation. Flow-GRPO also achieves substantial gains in human preference alignment. Notably, very little reward hacking occurred, meaning rewards did not increase at the cost of appreciable image quality or diversity degradation.

Results

TaskDatasetMetricValueModel
Image GenerationGenEvalOverall0.95SD3.5-Medium+Flow-GRPO
Text-to-Image GenerationGenEvalOverall0.95SD3.5-Medium+Flow-GRPO
10-shot image generationGenEvalOverall0.95SD3.5-Medium+Flow-GRPO
1 Image, 2*2 StitchiGenEvalOverall0.95SD3.5-Medium+Flow-GRPO

Related Papers

CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning2025-07-18fastWDM3D: Fast and Accurate 3D Healthy Tissue Inpainting2025-07-17Diffuman4D: 4D Consistent Human View Synthesis from Sparse-View Videos with Spatio-Temporal Diffusion Models2025-07-17Making Language Model a Hierarchical Classifier and Generator2025-07-17VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning2025-07-17Spectral Bellman Method: Unifying Representation and Exploration in RL2025-07-17Aligning Humans and Robots via Reinforcement Learning from Implicit Human Feedback2025-07-17VAR-MATH: Probing True Mathematical Reasoning in Large Language Models via Symbolic Multi-Instance Benchmarks2025-07-17