SyncAI.news, a Varaisys broadcasting
TyphoonWorld: A Residual Token World Model for Probabilistic Forecasting of Typhoon Imagery
XC

Xiaoyu Chen, Tianyi Li, Xuanyi Zhang

· 1 min read

ResearcharXiv cs.CV

TyphoonWorld: A Residual Token World Model for Probabilistic Forecasting of Typhoon Imagery

arXiv:2611.18740v1 Announce Type: new Abstract: A satellite forecast of a typhoon should give a low-error picture of the next two days, together with an ensemble whose members are coherent fields and whose spread responds to the storm at hand. A small direct regression network already gives the lowest squared error on held-out storms, so a generative model has to earn its cost through the part of the future that the regression smooths away. We present TyphoonWorld, a discrete-token world model that generates only that part. Conditioned on tokens of the regression forecast, the latest frames and the best-track state at forecast time, it samples residual fields lead by lead, and members add these residuals to the regression center. We compare it under one protocol with per-pixel Gaussian networks, a diffusion model of the same residuals, residuals from other storms and an unconditional token world model, on 276 forecast windows from 46 held-out storms of Digital Typhoon. TyphoonWorld matches the squared error of the regression, lowers the CRPS of the area colder than 220 K, a direct measure of deep convection, by 14 to 39% against every other ensemble, and ties its spread to forecast error as closely as the Gaussian network, with a correlation of 0.78 across windows. It uses 56% of the parameters of the unconditional world model.

Original source

This story was published by arXiv cs.CV and written by Xiaoyu Chen, Tianyi Li, Xuanyi Zhang. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News