Consistency Flow Matching: Defining Straight Flows with Velocity Consistency

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Ling, Zhang, Zixiang, Zhang, Zhilong, Liu, Xingchao, Xu, Minkai, Zhang, Wentao, Meng, Chenlin, Ermon, Stefano, Cui, Bin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909239378706432
author Yang, Ling
Zhang, Zixiang
Zhang, Zhilong
Liu, Xingchao
Xu, Minkai
Zhang, Wentao
Meng, Chenlin
Ermon, Stefano
Cui, Bin
author_facet Yang, Ling
Zhang, Zixiang
Zhang, Zhilong
Liu, Xingchao
Xu, Minkai
Zhang, Wentao
Meng, Chenlin
Ermon, Stefano
Cui, Bin
contents Flow matching (FM) is a general framework for defining probability paths via Ordinary Differential Equations (ODEs) to transform between noise and data samples. Recent approaches attempt to straighten these flow trajectories to generate high-quality samples with fewer function evaluations, typically through iterative rectification methods or optimal transport solutions. In this paper, we introduce Consistency Flow Matching (Consistency-FM), a novel FM method that explicitly enforces self-consistency in the velocity field. Consistency-FM directly defines straight flows starting from different times to the same endpoint, imposing constraints on their velocity values. Additionally, we propose a multi-segment training approach for Consistency-FM to enhance expressiveness, achieving a better trade-off between sampling quality and speed. Preliminary experiments demonstrate that our Consistency-FM significantly improves training efficiency by converging 4.4x faster than consistency models and 1.7x faster than rectified flow models while achieving better generation quality. Our code is available at: https://github.com/YangLing0818/consistency_flow_matching
format Preprint
id arxiv_https___arxiv_org_abs_2407_02398
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Consistency Flow Matching: Defining Straight Flows with Velocity Consistency
Yang, Ling
Zhang, Zixiang
Zhang, Zhilong
Liu, Xingchao
Xu, Minkai
Zhang, Wentao
Meng, Chenlin
Ermon, Stefano
Cui, Bin
Computer Vision and Pattern Recognition
Flow matching (FM) is a general framework for defining probability paths via Ordinary Differential Equations (ODEs) to transform between noise and data samples. Recent approaches attempt to straighten these flow trajectories to generate high-quality samples with fewer function evaluations, typically through iterative rectification methods or optimal transport solutions. In this paper, we introduce Consistency Flow Matching (Consistency-FM), a novel FM method that explicitly enforces self-consistency in the velocity field. Consistency-FM directly defines straight flows starting from different times to the same endpoint, imposing constraints on their velocity values. Additionally, we propose a multi-segment training approach for Consistency-FM to enhance expressiveness, achieving a better trade-off between sampling quality and speed. Preliminary experiments demonstrate that our Consistency-FM significantly improves training efficiency by converging 4.4x faster than consistency models and 1.7x faster than rectified flow models while achieving better generation quality. Our code is available at: https://github.com/YangLing0818/consistency_flow_matching
title Consistency Flow Matching: Defining Straight Flows with Velocity Consistency
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.02398