WxFM-XL: Adapting Univariate Foundation Models to Multi-Station Weather Forecasting

2026-10-07

AI Summary

This study addresses the limitation of univariate time series foundation models in multi-site meteorological forecasting, where spatial dependencies and station-specific error priors are often neglected. To this end, we propose a dynamic graph fusion framework. The core innovation lies in introducing the first cross-station error correlation prior graph and designing a dynamic fusion mechanism that leverages graph neural networks to adaptively integrate spatial topology with error distribution information, thereby effectively extending univariate foundation models to multi-site scenarios. Experimental results demonstrate that the proposed method significantly outperforms state-of-the-art baselines across multiple benchmark datasets.

Abstract

With the rise of univariate time series foundation models (e.g., Sundial, Timer), initial efforts have been made to extend them to multivariate settings. However, these models mainly focus on modeling correlations among variables. When they are applied to multi-station weather forecasting, two important factors are often overlooked: (1) the spatial information of stations, and (2) different error priors of different stations relative to the foundation model. In this paper, we propose WxFM-XL, a model for adapting univariate time series foundation models to multi-station weather forecasting. WxFM-XL introduces a cross-station error correlation prior graph to capture stationwise error priors with respect to the foundation model. Building on this, we further propose a dynamic fusion mechanism that adaptively integrates a spatial correlation graph with the error correlation prior graph. Experiments on multiple datasets demonstrate that our model outperforms state of the art baselines.

Problem

• univariate foundation models
• multi-station weather forecasting
• spatial information
• error priors
• time series

Innovation

• Univariate Foundation Models
• Multi-Station Weather Forecasting
• Error Correlation Prior Graph
• Dynamic Fusion Mechanism
• Spatial Correlation