Metraipolis › archive › dc_e72f59eb51e8
AREX: Affine-Residual Exponential Integrator for Few-Step Sampling in Flow Matching — arXiv:2610.03483
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03483
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:50:42.545Z
- SHA-256 of the archived text
da8413fce856a7b7f7e8c055a5ebc81637c2c03739356b2248e586ec801b9bd0- Stable URL of this page
https://metraipolis.com/documento/dc_e72f59eb51e8
How to cite
AREX: Affine-Residual Exponential Integrator for Few-Step Sampling in Flow Matching — arXiv:2610.03483 — arXiv, https://arxiv.org/abs/2610.03483 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e72f59eb51e8 (Metraipolis archive dc_e72f59eb51e8, sha256 da8413fce856a7b7f7e8c055a5ebc81637c2c03739356b2248e586ec801b9bd0)
BibTeX
@misc{dce72f59eb51e8,
title = {AREX: Affine-Residual Exponential Integrator for Few-Step Sampling in Flow Matching — arXiv:2610.03483},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03483},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e72f59eb51e8; sha256 da8413fce856a7b7f7e8c055a5ebc81637c2c03739356b2248e586ec801b9bd0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03483. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We introduce AREX, a training-free sampler for pretrained flow matching models that uses the target mean and covariance to capture an analytically tractable part of the sampling dynamics. We show that the velocity field of the moment-matched Gaussian target is the $L^2$-optimal affine approximation to the marginal velocity field. This motivates decomposition of the learned dynamics into an affine component over the whole sampling path, determined by the first two target moments, and a neural residual term. AREX keeps the affine component and integrates it using an explicit matrix-valued propagator. In turn, we only require to integrate over the residual term. This differs from scalar exponential integrators, which analytically handle only isotropic linear dynamics. Across image and text-to-image generation tasks, AREX consistently improves sample fidelity in the few-step sampling regime without retraining the underlying model. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (stat.ML) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.03483 — 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 2026-10-05T22:45:55.868Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).