stat.MLOct 30, 2026

Coarse-Graining Optimal Transport for Rollout-Free Post-Training of Flow Matching

Authors: Soon Hoe Lim, Aditi Gupta, Michael W. Mahoney, N. Benjamin Erichson

Organizations: Department of Mathematics, KTH Royal Institute of Technology · Nordita, KTH Royal Institute of Technology and Stockholm University · International Computer Science Institute · Department of Statistics, University of California at Berkeley · 

Abstract

Flow matching depends critically on the coupling used to pair source and target samples. Optimal transport (OT) couplings can reduce velocity conflict and straighten trajectories; but, in practice, challenges remain: the source comes from a continuous distribution, only a finite number of target samples data are accessible, and the neural network has finite capacity. For these reasons, making the coupling more OT-like may not consistently improve sampling quality, in particular in the low number-of-function-evaluation (NFE) regime. Motivated by this, we introduce ReCouple, a controllable framework that constructs coarse-grained OT couplings, to better understand rollout-free OT-based post-training. We also consider extensions of this basic framework: ReCouple+ extends this toward the continuous-source setting by increasing source resolution, converging to semi-discrete OT in the appropriate limit; and ReCouple++ further adds spatial regularization, motivated by the observation that straight trajectories need not correspond to spatially simple velocity fields. Empirical validation demonstrates the effectiveness of these methods, and it reveals an NFE-dependent interplay between coupling optimality, neural realizability, and generation quality.

Figures & tables

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Optimal Transport Flow Matching by Design

    Jun 2, 2026Shimon Malnick, Matan Rusanovsky, Ohad Fried +1Flow MatchingOne-Step Generative Modeling

  2. Beyond Transport Cost: Routing Differences between Flow Matching and Optimal Transport

    Oct 5, 2026Eungyeol Han, Jong-Seok LeeFlow MatchingFlow-Based Generative Modeling

  3. Scaling Online Optimal Transport Minibatches for Flow Matching

    Oct 13, 2026Peter Chen, Sergey LevineFlow MatchingFlow-Based Generative Modeling