General Proximal Flow Networks

Alexander Strunk Evercot AI Roland Assam Evercot AI · 2026

Read General Proximal Flow Networks by Alexander Strunk Evercot AI Roland Assam Evercot AI (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-00751.txt.

Description

Strunk and Assam generalise Bayesian Flow Networks into General Proximal Flow Networks, widening belief update rules. Where Bayesian updates equal proximal steps under Kullback-Leibler divergence, the new system accepts any such measure, such as Wasserstein distance, for step by step generative modelling. Training and sampling follow, linking the method to proximal optimisation and recovering standard updates as a special case. Tuning the measure to data geometry improves output quality.

Catalog record

CREATORAlexander Strunk Evercot AI Roland Assam Evercot AI
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE22688
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.00751

Browse the full library · About the Tower of Babel