发表机构
University of Technology Sydney; Tsinghua University; The Hong Kong University of Science and Technology (Guangzhou)（悉尼科技大学; 清华大学; 香港科技大学（广州））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对直接预测范式未显式约束跨变量结构的问题，提出跨变量损失CvLoss作为结构正则化项，可提升多变量时间序列预测模型性能，适配多种骨干网络。
AI 中文摘要
多变量时间序列预测存在独特挑战，因为未来变量常在共享系统动力学下协同演化。现有研究主要关注历史观测中的跨变量依赖，但未来值间的依赖探索甚少。现代预测模型大多遵循直接预测（Direct Forecasting, DF）范式，采用逐点目标生成多步预测，未显式约束跨变量结构。本研究表明，当存在跨变量和滞后依赖时，DF目标不匹配，存在目标缺口。为解决该问题，我们提出跨变量损失（Cross-Variable Loss, CvLoss），一种可插入的结构正则化项，用于约束跨变量图上的预测残差。CvLoss对预测片段上不一致的边级残差差异进行惩罚，鼓励同步和异步交互的一致性。实验显示，CvLoss可持续提升有竞争力的预测模型性能，优于代表性学习目标，且与多种预测骨干网络兼容。
英文摘要
Multivariate time series forecasting presents unique challenges because future variables often co-evolve under shared system dynamics. While existing studies mainly focus on cross-variable dependencies in historical observations, dependencies among future values are much less explored. Specifically, modern forecasting models largely follow the Direct Forecasting (DF) paradigm, generating multi-step forecasts with point-wise objectives that do not explicitly constrain cross-variable structure. In this work, we show that the DF objective is mismatched in the presence of cross-variable and lagged dependencies, revealing an objective gap. To address this issue, we propose \textbf{C}ross-\textbf{V}ariable \textbf{Loss} (CvLoss), a plug-in structural regularizer that constrains forecast residuals on a cross-variable graph. CvLoss penalizes inconsistent edge-wise residual differences over forecast patches, encouraging consistency across both synchronous and asynchronous interactions. Our experiments show that CvLoss consistently improves competitive forecasting models, outperforms representative learning objectives, and is compatible with a variety of forecasting backbones.
CommentsThis paper has been accepted by NeurIPS 2026
