AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

SynWeather: Weather Observation Data Synthesis Across Multiple Regions and Variables via a General Diffusion Transformer

Kaiyi Xu, Junchao Gong, Zhiwang Zhou, Zhangrui Li, Yuandong Pu, Yihao Liu, Ben Fei, Fenghua Ling, Wenlong Zhang, Lei Bai

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: Retrieval and matching with generative features · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use SynWeatherDiff instead of WeatherGFMgeneral model · CONUS CR synthesisRMSE: 3.124 -> 2.820Improved
use SynWeatherDiff instead of WeatherGFMgeneral model · CONUS precipitation synthesisCSI/15: 0.090 -> 0.113Improved
use SynWeatherDiff instead of UNet in pixel spacespecialized model · East Asia visible light synthesisSSIM: 0.917 -> 0.868Hurt
increase CONUS CR sampling ratio to 1/2 (ablation)SynWeatherDiff · CONUS precipitation synthesisCSI/15: 0.113 -> 0.137Improved
increase CONUS CR sampling ratio to 1/2 (ablation)SynWeatherDiff · CONUS CR synthesisRMSE: 2.82 -> 2.69Improved
increase CONUS precipitation sampling ratio to 1/2 (ablation)SynWeatherDiff · CONUS CR synthesisCSI/35: 0.158 -> 0.110Hurt
increase Europe visible light sampling ratio to 1/2 (ablation)SynWeatherDiff · Europe visible light synthesisSSIM: 0.864 -> 0.877Improved
increase MWBT sampling ratio to 1/2 (ablation)SynWeatherDiff · Europe visible light synthesisSSIM: 0.864 -> 0.879Improved
remove any input infrared channel group (SWIR, WV, LWIR, GAS) (ablation)SynWeatherDiff · six synthesis taskstask metrics: removing any group worsens all synthesis tasksImproved
use general text-prompted model instead of CONUS-trained specialized modelSynWeatherDiff · East Asia precipitation (OOD)RMSE: 1.383 -> 1.150Improved
use SynWeatherDiff instead of UNet and ViT in few-shot fine-tuningSynWeatherDiff · East Asia precipitation few-shotSSIM: 0.791 -> 0.828Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy