Consistency Flow Matching: Defining Straight Flows with Velocity Consistency

TL;DR

Consistency Flow Matching (Consistency-FM) introduces a self-consistent velocity field approach for defining straight probability flows, significantly improving training efficiency and sample quality in flow-based generative models.

Contribution

It proposes a novel self-consistent velocity field method for flow matching, enabling straight flows and faster training convergence.

Findings

Converges 4.4x faster than consistency models.

Converges 1.7x faster than rectified flow models.

Achieves better generation quality.

Abstract

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