Learning Straight Flows by Learning Curved Interpolants

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Shankar, Shiv, Geffner, Tomas
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909554164367360
author Shankar, Shiv
Geffner, Tomas
author_facet Shankar, Shiv
Geffner, Tomas
contents Flow matching models typically use linear interpolants to define the forward/noise addition process. This, together with the independent coupling between noise and target distributions, yields a vector field which is often non-straight. Such curved fields lead to a slow inference/generation process. In this work, we propose to learn flexible (potentially curved) interpolants in order to learn straight vector fields to enable faster generation. We formulate this via a multi-level optimization problem and propose an efficient approximate procedure to solve it. Our framework provides an end-to-end and simulation-free optimization procedure, which can be leveraged to learn straight line generative trajectories.
format Preprint
id arxiv_https___arxiv_org_abs_2503_20719
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Straight Flows by Learning Curved Interpolants
Shankar, Shiv
Geffner, Tomas
Machine Learning
Flow matching models typically use linear interpolants to define the forward/noise addition process. This, together with the independent coupling between noise and target distributions, yields a vector field which is often non-straight. Such curved fields lead to a slow inference/generation process. In this work, we propose to learn flexible (potentially curved) interpolants in order to learn straight vector fields to enable faster generation. We formulate this via a multi-level optimization problem and propose an efficient approximate procedure to solve it. Our framework provides an end-to-end and simulation-free optimization procedure, which can be leveraged to learn straight line generative trajectories.
title Learning Straight Flows by Learning Curved Interpolants
topic Machine Learning
url https://arxiv.org/abs/2503.20719