Graph Flow Matching: Enhancing Image Generation with Neighbor-Aware Flow Fields

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Siddiqui, Md Shahriar Rahim, Eliasof, Moshe, Haber, Eldad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911256568397824
author Siddiqui, Md Shahriar Rahim
Eliasof, Moshe
Haber, Eldad
author_facet Siddiqui, Md Shahriar Rahim
Eliasof, Moshe
Haber, Eldad
contents Flow matching casts sample generation as learning a continuous-time velocity field that transports noise to data. Existing flow matching networks typically predict each point's velocity independently, considering only its location and time along its flow trajectory, and ignoring neighboring points. However, this pointwise approach may overlook correlations between points along the generation trajectory that could enhance velocity predictions, thereby improving downstream generation quality. To address this, we propose Graph Flow Matching (GFM), a lightweight enhancement that decomposes the learned velocity into a reaction term -- any standard flow matching network -- and a diffusion term that aggregates neighbor information via a graph neural module. This reaction-diffusion formulation retains the scalability of deep flow models while enriching velocity predictions with local context, all at minimal additional computational cost. Operating in the latent space of a pretrained variational autoencoder, GFM consistently improves Fréchet Inception Distance (FID) and recall across five image generation benchmarks (LSUN Church, LSUN Bedroom, FFHQ, AFHQ-Cat, and CelebA-HQ at $256\times256$), demonstrating its effectiveness as a modular enhancement to existing flow matching architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2505_24434
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph Flow Matching: Enhancing Image Generation with Neighbor-Aware Flow Fields
Siddiqui, Md Shahriar Rahim
Eliasof, Moshe
Haber, Eldad
Machine Learning
Computer Vision and Pattern Recognition
Flow matching casts sample generation as learning a continuous-time velocity field that transports noise to data. Existing flow matching networks typically predict each point's velocity independently, considering only its location and time along its flow trajectory, and ignoring neighboring points. However, this pointwise approach may overlook correlations between points along the generation trajectory that could enhance velocity predictions, thereby improving downstream generation quality. To address this, we propose Graph Flow Matching (GFM), a lightweight enhancement that decomposes the learned velocity into a reaction term -- any standard flow matching network -- and a diffusion term that aggregates neighbor information via a graph neural module. This reaction-diffusion formulation retains the scalability of deep flow models while enriching velocity predictions with local context, all at minimal additional computational cost. Operating in the latent space of a pretrained variational autoencoder, GFM consistently improves Fréchet Inception Distance (FID) and recall across five image generation benchmarks (LSUN Church, LSUN Bedroom, FFHQ, AFHQ-Cat, and CelebA-HQ at $256\times256$), demonstrating its effectiveness as a modular enhancement to existing flow matching architectures.
title Graph Flow Matching: Enhancing Image Generation with Neighbor-Aware Flow Fields
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.24434