LLM增强时空学习预测网格级共享单车需求
原标题：LLM-enhanced spatio-temporal learning for grid-level docked bike sharing demand prediction
AI 摘要
该研究提出一种LLM增强的时空学习框架，用于网格级有桩共享单车需求预测。LLM将城市事件文本（市政活动、新闻、交通公告）解析为强度、空间影响半径和时间滞后三个物理参数，构建高斯衰减场并注入零膨胀自适应时空图卷积网络（ZI-ASTGCN）。在巴塞罗那Bicing数据集上，该方法优于现有神经基线，尤其在高需求时段提升明显。
正文节选
LLM-Enhanced Spatio-temporal Learning for Grid-Level Docked Bike Sharing Demand Prediction Abstract Short-term bike-sharing demand forecasting is complicated by spatial-temporal non-stationarity and the practical difficulty of incorporating unstructured external text into numerical pipelines. Conventional approaches rely on historical flow sequences and fixed graph structures, thereby constraining their accuracy when anomalous social events perturb normal travel patterns. We propose a forecastin