用平均速度建模生成过程，单次推断即达顶尖性能。
Mean Flows for One-step Generative Modeling
- 以平均速度替代瞬时速度构建生成流程，理论更扎实。
- 在ImageNet上仅需1次函数求值（1-NFE），FID达3.43。
- 无需预训练或教学蒸馏，适合追求高效生成的研究者。
我们提出一种原理严谨且高效的单步生成建模框架。与以往基于瞬时速度的流匹配方法不同，我们引入平均速度来刻画流场，并推导出平均速度与瞬时速度之间的明确对应关系，用于指导神经网络训练。所提出的MeanFlow模型自洽完整，无需预训练、蒸馏或课程学习。实验表明，该方法在从零训练的ImageNet 256x256数据集上仅通过一次函数评估（1-NFE）即达到FID 3.43，显著优于此前最优的单步扩散/流模型。本研究大幅缩小了单步与多步生成模型间的性能差距，有望推动对这类强大模型基础理论的重新审视。
原文摘要 · Abstract (English)
We propose a principled and effective framework for one-step generative modeling. We introduce the notion of average velocity to characterize flow fields, in contrast to instantaneous velocity modeled by Flow Matching methods. A well-defined identity between average and instantaneous velocities is derived and used to guide neural network training. Our method, termed the MeanFlow model, is self-contained and requires no pre-training, distillation, or curriculum learning. MeanFlow demonstrates strong empirical performance: it achieves an FID of 3.43 with a single function evaluation (1-NFE) on ImageNet 256x256 trained from scratch, significantly outperforming previous state-of-the-art one-step diffusion/flow models. Our study substantially narrows the gap between one-step diffusion/flow models and their multi-step predecessors, and we hope it will motivate future research to revisit the foundations of these powerful models.
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。