paper-with-me

Papers

Flow-Map GRPO: Reinforcement Learning for Few-Step Flow-Map Generators via Anchored Stochastic Composition

2026-07-01 · Zhiqi Li, Wen Zhang, Bo Zhu arxiv

Few-step flow-map generators, such as consistency models and MeanFlow, accelerate sampling by directly learning long-range transport maps between noise and data. However, these models are typically deterministic, which makes them difficult to optimize with reinforcement learning (RL) post-training methods that require stochastic trajectories and well-defined likelihood ratios. Existing SDE-based stochasticization techniques are designed for velocity-based samplers with infinitesimal or finely discretized transitions, and therefore do not directly apply to long-range flow maps. In this work, we propose Flow-Map GRPO, an online RL post-training framework for deterministic few-step flow-map generators. The key component is Anchored Stochastic Flow Map Composition (ASFMC), a path-preserving stochasticization mechanism that introduces randomness through anchor-based conditional resampling while preserving the original marginal probability path of the deterministic flow map. We derive GRPO objectives for both single-time and two-time flow-map parameterizations. Experiments on few-step FLUX-based text-to-image generators, including MeanFlow and sCM, show that Flow-Map GRPO improves pretrained deterministic flow-map models across reward-based, perceptual, and task-level evaluation metrics. Our results demonstrate that deterministic few-step flow-map generators can be effectively aligned with RL post-training without modifying their original model parameterization or retraining them as native stochastic models.

📄 PDF Abstract BibTeX arXiv:2607.00535

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Faster and Better Alignment for Flow Matching Models via Step-aware Advantages

2026-02-02 · Zhixiong Yue, Zixuan Ni, Feiyang Ye, Jinshan Zhang 외 arxiv

Recent advances in flow matching models, particularly with reinforcement learning (RL), have significantly enhanced human preference alignment in few-step text-to-image generators. However, existing RL-based approaches f…

Reinforcement LearningImage Generation

Reinforcing Few-step Generators via Reward-Tilted Distribution Matching

2026-05-25 · Yushi Huang, Xiangxin Zhou, Ruoyu Wang, Chi Zhang 외 arxiv

Recent advances in few-step diffusion distillation have enabled efficient image generation, yet aligning these models with human preferences remains challenging. We propose Reward-Tilted Distribution Matching Distillatio…

Text-to-Image GenerationReinforcement Learning

Stepwise Credit Assignment for GRPO on Flow-Matching Models

2026-03-30 · Yash Savani, Branislav Kveton, Yuchen Liu, Yilin Wang 외 arxiv

Flow-GRPO successfully applies reinforcement learning to flow models, but uses uniform credit assignment across all steps. This ignores the temporal structure of diffusion generation: early steps determine composition an…

Reinforcement Learning

Flow-GRPO: Training Flow Matching Models via Online RL

2025-05-08 · Jie Liu, Gongye Liu, Jiajun Liang, Yangguang Li 외

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…

DenoisingDiversityReinforcement Learning (RL)Text Generation+1

LC-GRPO: Bridging Train-Inference Gap for Flow-Based GRPO with Langevin Correction

2026-08-06 · Yingqing Guo, Hui Yuan, Zijian He, Mengdi Wang 외 arxiv

Flow-based generative models are typically sampled by solving a deterministic ordinary differential equation (ODE), whereas online reinforcement learning requires stochastic rollouts for policy exploration and optimizati…

Reinforcement Learning