Metraipolis › archive › dc_d2eb28edeae8

Coarse-Graining Optimal Transport for Rollout-Free Post-Training of Flow Matching — arXiv:2610.41269

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41269
License (as archived)
CC0-1.0
Archived
2026-11-02T04:53:14.103Z
SHA-256 of the archived text
1596bbaf896a2729fd169365f49c83beb2475eaaab5882e1c0cdc4242a155361
Stable URL of this page
https://metraipolis.com/documento/dc_d2eb28edeae8

How to cite

Coarse-Graining Optimal Transport for Rollout-Free Post-Training of Flow Matching — arXiv:2610.41269 — arXiv (en), https://arxiv.org/abs/2610.41269 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2eb28edeae8 (Metraipolis archive dc_d2eb28edeae8, sha256 1596bbaf896a2729fd169365f49c83beb2475eaaab5882e1c0cdc4242a155361)
BibTeX
@misc{dcd2eb28edeae8,
  title = {Coarse-Graining Optimal Transport for Rollout-Free Post-Training of Flow Matching — arXiv:2610.41269},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41269},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2eb28edeae8; sha256 1596bbaf896a2729fd169365f49c83beb2475eaaab5882e1c0cdc4242a155361},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41269. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.41269 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-02T04:45:12.950Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).