Consistency Flow Matching: Defining Straight Flows with Velocity Consistency
This work addresses a bottleneck in generative modeling for researchers and practitioners by providing a more efficient method for high-quality sample generation, though it is incremental as it builds on existing flow matching frameworks.
The paper tackles the problem of improving sampling efficiency and quality in flow matching by introducing Consistency Flow Matching, which enforces velocity consistency to define straight flows, resulting in 4.4x faster convergence than consistency models and 1.7x faster than rectified flow models with better generation quality.
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