STCast：面向全球与区域天气预报的自适应边界对齐
STCast: Adaptive Boundary Alignment for Global and Regional Weather Forecasting
Spatial-Aligned Attention 用注意力对齐全局与区域空间分布并据此自适应细化区域边界，这一「注意力导出对齐模式→动态边界」的思路可迁移到遥感/气象等大场景中的区域裁剪与感兴趣区域自适应划分；TMoE 按离散高斯分布路由不同月份变量的做法，也为时序视觉模型中的条件专家路由提供参考。
原文依据 employs a Spatial-Aligned Attention (SAA) mechanism, which aligns global and regional spatial distributions to initialize boundaries and adaptively refines them
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论为获得更精细的区域预报，许多工作探索了从全球大气中进行区域集成，例如在物理方法中求解边界方程，或在数据驱动方法中从全球预报中裁剪区域。然而，这些方法的有效性往往受限于静态且不精确的区域边界，导致泛化能力较差。为解决该问题，我们提出空间-时序天气预报（STCast），一个用于自适应区域边界优化与动态月度预报分配的新型 AI 驱动框架。具体而言，我们的方法采用空间对齐注意力（SAA）机制，该机制对齐全球与区域的空间分布以初始化边界，并基于注意力导出的对齐模式对其进行自适应细化。此外，我们设计了时序混合专家（TMoE）模块，其中来自不同月份的大气变量通过离散高斯分布被动态路由到专门的专家，从而增强模型捕获时序模式的能力。除全球与区域预报外，STCast 还在极端事件预测与集合预报上进行了评估。实验结果表明，在全部四项任务上该方法均持续优于其他方法。代码：https://github.com/chenhao-zju/STCast
英文原文摘要 1338 字符
To gain finer regional forecasts, many works have explored the regional integration from the global atmosphere, e.g., by solving boundary equations in physics-based methods or cropping regions from global forecasts in data-driven methods. However, the effectiveness of these methods is often constrained by static and imprecise regional boundaries, resulting in poor generalization ability. To address this issue, we propose Spatial-Temporal Weather Forecasting (STCast), a novel AI-driven framework for adaptive regional boundary optimization and dynamic monthly forecast allocation. Specifically, our approach employs a Spatial-Aligned Attention (SAA) mechanism, which aligns global and regional spatial distributions to initialize boundaries and adaptively refines them based on attention-derived alignment patterns. Furthermore, we design a Temporal Mixture-of-Experts (TMoE) module, where atmospheric variables from distinct months are dynamically routed to specialized experts using a discrete Gaussian distribution, enhancing the model's ability to capture temporal patterns. Beyond global and regional forecasting, STCast is evaluated on extreme event prediction and ensemble forecasting. Experimental results demonstrate consistent superiority over other methods across all four tasks. Code: https://github.com/chenhao-zju/STCast