Saved in:
Bibliographic Details
Main Authors: Shankar, Shiv, Geffner, Tomas
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.20719
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of 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.