统一强化学习框架，让流匹配模型更易对齐人类偏好。
Flow-Factory: A Unified Framework for Reinforcement Learning in Flow-Matching Models
- 模块化架构解耦算法、模型与奖励，支持灵活扩展。
- 兼容GRPO、DiffusionNFT等多算法，适配多种主流模型。
- 提供内存优化与分布式训练，适合快速实验与规模化研究。
强化学习已成为对齐扩散模型与流匹配模型与人类偏好的有前景范式，但实践者常面临代码库零散、模型专用实现及工程复杂性问题。我们提出Flow-Factory，一个通过模块化注册表架构解耦算法、模型与奖励的统一框架。该设计支持新算法与架构的无缝集成，已验证可兼容GRPO、DiffusionNFT与AWM，并在Flux、Qwen-Image及WAN视频模型上实现部署。通过降低实现开销，Flow-Factory助力研究者轻松快速原型设计与规模化创新。框架提供生产就绪的内存优化、灵活的多奖励训练以及无缝分布式训练支持。代码已开源：https://github.com/X-GenGroup/Flow-Factory。
原文摘要 · Abstract (English)
Reinforcement learning has emerged as a promising paradigm for aligning diffusion and flow-matching models with human preferences, yet practitioners face fragmented codebases, model-specific implementations, and engineering complexity. We introduce Flow-Factory, a unified framework that decouples algorithms, models, and rewards through through a modular, registry-based architecture. This design enables seamless integration of new algorithms and architectures, as demonstrated by our support for GRPO, DiffusionNFT, and AWM across Flux, Qwen-Image, and WAN video models. By minimizing implementation overhead, Flow-Factory empowers researchers to rapidly prototype and scale future innovations with ease. Flow-Factory provides production-ready memory optimization, flexible multi-reward training, and seamless distributed training support. The codebase is available at https://github.com/X-GenGroup/Flow-Factory.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。