Towards Scalable One-Step Generative Modeling for Autoregressive Dynamical System Forecasting
Tianyue Yang, Xiao Xue · 2026
Read Towards Scalable One-Step Generative Modeling for Autoregressive Dynamical System Forecasting by Tianyue Yang, Xiao Xue (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-2605-05540.txt.
Description
Predicting turbulent flows with AI surrogates is easy for a few steps and hard for long horizons. Tianyue Yang and Xiao Xue present MeLISA, a generative model built on pixel-space MeanFlow. Neural operators drift over long rollouts, while diffusion models need many slow steps for each run.
MeLISA generates each block of the future in a single model evaluation. Two consistency losses keep long paths honest, guarding energy spectra and mixing behavior. On high-resolution turbulence benchmarks it beat neural-operator baselines on short-term accuracy and long-run statistics, at comparable or better speed.
Catalog record
| CREATOR | Tianyue Yang, Xiao Xue |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 93 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.05540