Benchmarking Long Roll-outs of Auto-regressive Neural Operators for the Compressible Navier-Stokes Equations with Conserved Quantity Correction

Sean F. Current and Chandan Kumar and Datta V. Gaitonde and Srinivasan Parthasarathy · 2026

Read Benchmarking Long Roll-outs of Auto-regressive Neural Operators for the Compressible Navier-Stokes Equations with Conserved Quantity Correction by Sean F. Current and Chandan Kumar and Datta V. Gaitonde and Srinivasan Parthasarathy (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-2601-22541.txt.

Description

Sean Current, Chandan Kumar, Datta Gaitonde, and Srinivasan Parthasarathy test how far neural surrogates can roll out a hard airflow on their own. Neural operators learn to step a physics field forward at great speed. Their 2026 preprint uses the compressible Navier-Stokes equations, the core rules of gas flow, with shocks and sharp fronts.

Long rollouts are where such tools tend to drift or blow up. The bench marks that drift over many steps. Honest limits help engineers know when a fast surrogate can be trusted. The work serves flight and turbine design.

Catalog record

CREATORSean F. Current and Chandan Kumar and Datta V. Gaitonde and Srinivasan Parthasarathy
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-22541

Browse the full library · About the Tower of Babel