WxFM-XL: Adapting Univariate Foundation Models to Multi-Station Weather Forecasting
WxFM-XL adapts univariate weather foundation models for multi-station forecasting by decomposing forecasts into low, mid, and high frequency components. It introduces a dynamic fusion mechanism that blends static spatial correlations with frequency-specific error correlation priors to refine predictions. This results in state-of-the-art performance across various datasets.