← Back to all news
arXiv Machine Learning By Zander W. Blasingame, Antonio Franca, Raul Minan, Christian Koke, Niklas Rindtorff, Alexander Tong

Improved Training of Strong Stochastic Flow Maps

Read the original on arXiv Machine Learning →

The paper introduces methods to improve training of stochastic flow maps for stochastic differential equations. It proposes representing Brownian paths with a shifted Legendre polynomial basis and using a Wong‑Zakai styled objective to enable derivative‑based losses. The authors also provide a fine‑tuning recipe that turns large pre‑trained flow models into strong stochastic flow maps, achieving a three‑fold improvement on ImageNet compared to SSFM.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
Jun 2

Strong Stochastic Flow Maps

arXiv:2606. 01086v1 Announce Type: cross Abstract: Flow and diffusion models generate high-quality samples in many modalities; however, many network evaluations are required during inference due to numerical integration of an underlying differential equation.

By Sam McCallum, Zander W. Blasingame, Timothy Herschell, Niklas Rindtorff, Alexander Tong, James Foster
diffusion
More like this →
arXiv Machine Learning
Jun 10

It\^o maps for any-step SDEs

arXiv:2606. 11156v1 Announce Type: cross Abstract: Recent one-step generative models accelerate sampling by learning deterministic flow maps of the underlying dynamics.

By Zhengkai Pan, Peter Potaptchik, Wenxi Yao, Michael S. Albergo, Jakiw Pidstrigach
efficiencybenchmarks
More like this →
arXiv Machine Learning
Oct 16

On the Stability and Convergence of Flow Map Training

The paper investigates the training dynamics of flow map models, which convert diffusion trajectories into one‑step samplers for efficient probabilistic generation. It provides a theoretical analysis of the functional gradient flow induced by MeanFlow and Lagrangian losses, proving their convergence to the true flow map and identifying the risk of unbounded updates. The authors show that adaptive normalization and a specific reweighting scheme not only bound updates but also guarantee convergence to a neighborhood of the solution, and they confirm these findings with synthetic experiments and image‑generation benchmarks.

By Ishan Nath, Tom Stanic, Avishek Joey Bose, Gauthier Gidel, Quentin Bertrand
diffusion
More like this →