Saved in:
Bibliographic Details
Main Authors: Zhou, Linqi, Ermon, Stefano, Song, Jiaming
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.07565
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Diffusion models and Flow Matching generate high-quality samples but are slow at inference, and distilling them into few-step models often leads to instability and extensive tuning. To resolve these trade-offs, we propose Inductive Moment Matching (IMM), a new class of generative models for one- or few-step sampling with a single-stage training procedure. Unlike distillation, IMM does not require pre-training initialization and optimization of two networks; and unlike Consistency Models, IMM guarantees distribution-level convergence and remains stable under various hyperparameters and standard model architectures. IMM surpasses diffusion models on ImageNet-256x256 with 1.99 FID using only 8 inference steps and achieves state-of-the-art 2-step FID of 1.98 on CIFAR-10 for a model trained from scratch.