标签
论文提出 FireGraphNet,将12小时野火蔓延预测建模为有向 H3 图上的节点分类任务,在加州测试火灾上平均精度达到最强距离基线的1.7–1.9倍,并分析了图消息传递对预测能力的贡献。
M2Weather 提出了一个多站点、多变量联合天气预报基准,涵盖法国、欧洲和全球尺度共 2,809 个观测站点,并设计了一种即插即用的适配器,通过建模站点与变量之间的关系来提升预报效果。
This paper introduces GeoID-PINN, a physics-informed neural network for regional epidemic inference that incorporates geographic coupling via a spatial source-composition matrix. It demonstrates that accurate trajectory forecasts do not guarantee correct recovery of regional dependence structures, using simulations and COVID-19 data from Louisiana counties.