Path-Guided Flow Matching for Dataset Distillation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Xuhui, Luo, Zhengquan, Liu, Xiwei, Yu, Yongqiang, Xu, Zhiqiang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910012892250112
author Li, Xuhui
Luo, Zhengquan
Liu, Xiwei
Yu, Yongqiang
Xu, Zhiqiang
author_facet Li, Xuhui
Luo, Zhengquan
Liu, Xiwei
Yu, Yongqiang
Xu, Zhiqiang
contents Dataset distillation compresses large datasets into compact synthetic sets with comparable performance in training models. Despite recent progress on diffusion-based distillation, this type of method typically depends on heuristic guidance or prototype assignment, which comes with time-consuming sampling and trajectory instability and thus hurts downstream generalization especially under strong control or low IPC. We propose \emph{Path-Guided Flow Matching (PGFM)}, the first flow matching-based framework for generative distillation, which enables fast deterministic synthesis by solving an ODE in a few steps. PGFM conducts flow matching in the latent space of a frozen VAE to learn class-conditional transport from Gaussian noise to data distribution. Particularly, we develop a continuous path-to-prototype guidance algorithm for ODE-consistent path control, which allows trajectories to reliably land on assigned prototypes while preserving diversity and efficiency. Extensive experiments across high-resolution benchmarks demonstrate that PGFM matches or surpasses prior diffusion-based distillation approaches with fewer steps of sampling while delivering competitive performance with remarkably improved efficiency, e.g., 7.6$\times$ more efficient than the diffusion-based counterparts with 78\% mode coverage.
format Preprint
id arxiv_https___arxiv_org_abs_2602_05616
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Path-Guided Flow Matching for Dataset Distillation
Li, Xuhui
Luo, Zhengquan
Liu, Xiwei
Yu, Yongqiang
Xu, Zhiqiang
Machine Learning
Artificial Intelligence
Dataset distillation compresses large datasets into compact synthetic sets with comparable performance in training models. Despite recent progress on diffusion-based distillation, this type of method typically depends on heuristic guidance or prototype assignment, which comes with time-consuming sampling and trajectory instability and thus hurts downstream generalization especially under strong control or low IPC. We propose \emph{Path-Guided Flow Matching (PGFM)}, the first flow matching-based framework for generative distillation, which enables fast deterministic synthesis by solving an ODE in a few steps. PGFM conducts flow matching in the latent space of a frozen VAE to learn class-conditional transport from Gaussian noise to data distribution. Particularly, we develop a continuous path-to-prototype guidance algorithm for ODE-consistent path control, which allows trajectories to reliably land on assigned prototypes while preserving diversity and efficiency. Extensive experiments across high-resolution benchmarks demonstrate that PGFM matches or surpasses prior diffusion-based distillation approaches with fewer steps of sampling while delivering competitive performance with remarkably improved efficiency, e.g., 7.6$\times$ more efficient than the diffusion-based counterparts with 78\% mode coverage.
title Path-Guided Flow Matching for Dataset Distillation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.05616