PulseAugur
实时 18:29:36
English(EN) LC-GRPO: Bridging Train-Inference Gap for Flow-Based GRPO with Langevin Correction

LC-GRPO框架通过Langevin校正改进生成模型训练

研究人员推出了一种新颖的基于流的GRPO框架LC-GRPO,该框架结合了Langevin校正,以弥合生成模型中训练和推理之间的差距。该方法解决了在使用随机微分方程(SDE)进行训练和使用常微分方程(ODE)进行推理时出现的差异,这种差异可能导致样本模糊和性能下降。通过在ODE欧拉步之后应用Langevin校正,LC-GRPO在理论上减少了Wasserstein误差,并与标准的SDE离散化相比提高了样本精度。在SD3.5-Medium、FLUX.1-Dev和HunyuanVideo等模型上的实验表明,在文本到图像和文本到视频生成任务的奖励优化方面取得了持续改进,同时保持了生成质量。 AI

影响 这项研究提供了一种方法来增强基于流的生成模型在训练和推理之间的一致性,从而可能提高它们在文本到图像和文本到视频生成等任务中的性能。

排序理由 该集群描述了一篇详细介绍改进生成模型新方法的最新研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

LC-GRPO框架通过Langevin校正改进生成模型训练

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群描述了一篇详细介绍改进生成模型新方法的最新研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
48 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Yingqing Guo, Hui Yuan, Zijian He, Mengdi Wang, Zheng Ding ·

    LC-GRPO:利用Langevin校正弥合基于流的GRPO的训练-推理鸿沟

    arXiv:2608.05600v1 Announce Type: cross Abstract: 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 optimization. Existi…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    LC-GRPO:为具有 Langevin 校正的基于流的 GRPO 弥合训练-推理差距

    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 optimization. Existing GRPO methods for flow models therefore replace …