AI Paper Atlas

Research map / Probabilistic and scientific modeling / Differential equations and PDEs

FuXi-RTM: A Physics-Guided Prediction Framework with Radiative Transfer Modeling

Qiusheng Huang, Xiaohui Zhong, Xu Fan, Hao Li

ICCV 2025 · Probabilistic and scientific modeling › Differential equations and PDEs · approach: Physics-informed operator learning

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
add DLRTM radiative flux regularizationFuXi-base · ERA5 3320 var-lead combosfraction of combos improved: 88.51% of combinations improvedImproved
add DLRTM radiative flux regularizationFuXi-base · ERA5, CCfraction of leads improved: 95.38%Improved
add DLRTM radiative flux regularizationFuXi-base · ERA5, FALnRMSE: exceeding 7%Improved
add DLRTM regularizationFuXi-base · ERA5, Q50RMSE: 0.1735 -> 0.1546Improved
add DLRTM regularizationFuXi-base · ERA5, TPRMSE: 2.4127 -> 2.4092Improved
add DLRTM regularizationFuXi-base · ERA5, Q at 1000 hPaRMSE: trails for up to 1.25 daysMixed
add sunlit region-centered sampling instead of global random sampling (ablation)FuXi-RTM · ERA5, Q50RMSE: 0.1857 -> 0.1546Improved
add sunlit region-centered sampling instead of global random sampling (ablation)FuXi-RTM · ERA5, FALRMSE: 0.02515 -> 0.02300Improved
supervise only surface SW fluxes instead of all 13 levels SW+LW (ablation)FuXi-RTM · ERA5, Q50RMSE: 0.1697 -> 0.1546Improved
supervise only surface SW fluxes instead of all 13 levels SW (ablation)FuXi-RTM · ERA5, TTRRMSE: 158.4191 -> 158.1290Improved
supervise raw SW fluxes instead of net surface SW (GSW) (ablation)FuXi-RTM · ERA5, Q500RMSE: 0.7567 -> 0.7388Improved
supervise raw SW fluxes instead of ISSRD (ablation)FuXi-RTM · ERA5, Q50RMSE: 0.1741 -> 0.1546Improved

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