发表机构
University of Virginia; Datadog; Florida State University; Northeastern University（弗吉尼亚大学; Datadog; 佛罗里达州立大学; 东北大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有时空预测模型忽视数据耦合模式差异的问题，提出AdaST框架，通过分解-重组自适应调节时空建模，显著超越现有基线。
AI 中文摘要
时空（ST）预测支撑着许多现实世界系统，如交通、气候和能源网络。尽管现有方法隐式假设强时空耦合，我们观察到现实世界的时空数据表现出不同的耦合机制，范围从时间主导、空间主导到强耦合模式。这种不匹配导致当前模型在一种相关性占主导时遭受虚假依赖和性能下降。为克服这一局限，我们旨在基于数据固有的耦合结构动态调节时空建模。然而，存在三个关键挑战：未知的耦合结构、异质的耦合动态和次优的空间建模。我们提出AdaST，一种自适应时空预测框架，通过分解-重组范式应对这些挑战。AdaST使用异质性感知专家将输入分解为捕捉不同耦合模式的组件。每个组件由角色对齐的模块处理，一个相关性感知的自适应重组器整合它们以进行最终预测。大量实验证实AdaST显著优于最先进的基线，验证了自适应方法的必要性。
英文摘要
Spatial-temporal (ST) forecasting underpins many real-world systems such as traffic, climate, and energy networks. While existing methods implicitly assume strong spatiotemporal coupling, we observe that real-world ST data exhibits distinct coupling regimes, ranging from temporal-dominated and spatial-dominated to strongly coupled patterns. This mismatch causes current models to suffer from spurious dependencies and degraded performance when one correlation dominates. To overcome this limitation, we aim to dynamically modulate spatial and temporal modeling based on the data's inherent coupling structure. However, three key challenges exist: unknown coupling structure, heterogeneous coupling dynamics, and suboptimal spatial modeling. We propose AdaST, an adaptive ST forecasting framework that tackles these challenges through a decompose-recompose paradigm. AdaST factorizes inputs into components capturing different coupling patterns using heterogeneity-aware experts. Each component is processed by role-aligned modules, and a correlation-informed adaptive recomposer integrates them for final prediction. Extensive experiments confirm that AdaST significantly outperforms state-of-the-art baselines, validating the necessity of an adaptive approach.
CommentsAccepted to NeurIPS 2026 (main conference). Code: https://github.com/LzyFischer/AdaST
