vix.ing · top · new · best · stats · spec

Mean Flows for One-step Generative Modeling

2025/05/19 by Zhengyang Geng, Mingyang Deng, Geng, Zhengyang +7 · 3 voices · 147 citations
Computer Science · Engineering · Physics and Astronomy · #3D Shape Modeling and Analysis #Generative Adversarial Networks and Image Synthesis #Model Reduction and Neural Networks #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2505.13447

openalex publication_date 2025/05/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

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.

Cited by

Discussions

Related