Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schusterbauer, Johannes, Gui, Ming, Fundel, Frank, Ommer, Björn
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912410880704512
author Schusterbauer, Johannes
Gui, Ming
Fundel, Frank
Ommer, Björn
author_facet Schusterbauer, Johannes
Gui, Ming
Fundel, Frank
Ommer, Björn
contents Diffusion models have revolutionized generative tasks through high-fidelity outputs, yet flow matching (FM) offers faster inference and empirical performance gains. However, current foundation FM models are computationally prohibitive for finetuning, while diffusion models like Stable Diffusion benefit from efficient architectures and ecosystem support. This work addresses the critical challenge of efficiently transferring knowledge from pre-trained diffusion models to flow matching. We propose Diff2Flow, a novel framework that systematically bridges diffusion and FM paradigms by rescaling timesteps, aligning interpolants, and deriving FM-compatible velocity fields from diffusion predictions. This alignment enables direct and efficient FM finetuning of diffusion priors with no extra computation overhead. Our experiments demonstrate that Diff2Flow outperforms naïve FM and diffusion finetuning particularly under parameter-efficient constraints, while achieving superior or competitive performance across diverse downstream tasks compared to state-of-the-art methods. We will release our code at https://github.com/CompVis/diff2flow.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02221
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment
Schusterbauer, Johannes
Gui, Ming
Fundel, Frank
Ommer, Björn
Computer Vision and Pattern Recognition
Machine Learning
Diffusion models have revolutionized generative tasks through high-fidelity outputs, yet flow matching (FM) offers faster inference and empirical performance gains. However, current foundation FM models are computationally prohibitive for finetuning, while diffusion models like Stable Diffusion benefit from efficient architectures and ecosystem support. This work addresses the critical challenge of efficiently transferring knowledge from pre-trained diffusion models to flow matching. We propose Diff2Flow, a novel framework that systematically bridges diffusion and FM paradigms by rescaling timesteps, aligning interpolants, and deriving FM-compatible velocity fields from diffusion predictions. This alignment enables direct and efficient FM finetuning of diffusion priors with no extra computation overhead. Our experiments demonstrate that Diff2Flow outperforms naïve FM and diffusion finetuning particularly under parameter-efficient constraints, while achieving superior or competitive performance across diverse downstream tasks compared to state-of-the-art methods. We will release our code at https://github.com/CompVis/diff2flow.
title Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2506.02221