用双代理序列让单变量模型高效预测多变量时间序列。
DualWeaver: Synergistic Feature Weaving Surrogates for Multivariate Forecasting with Univariate Time Series Foundation Models
- 设计对称双代理序列，融合变量间依赖关系。
- 在多个真实数据集上超越现有方法，精度与稳定性更优。
- 无需额外解码参数，适合快速部署的多变量预测场景。
时间序列基础模型（TSFMs）通过大规模预训练在单变量预测中表现优异，但将其成功扩展至多变量预测仍具挑战。为此，我们提出 DualWeaver 框架，通过一对可学习、结构对称的代理序列，将单变量时间序列基础模型（Uni-TSFMs）适配于多变量预测。这些代理序列由共享的辅助特征融合模块生成，捕捉变量间依赖关系，并通过预测目标映射为符合 TSFM 输入格式的序列。对称结构支持无参重建最终预测结果，无需额外参数化解码。进一步引入理论支撑的正则化项，提升对适应崩溃的鲁棒性。在多种真实世界数据集上的实验表明，DualWeaver 在精度和稳定性上均优于当前最优的多变量预测方法。代码已开源：https://github.com/li-jinpeng/DualWeaver。
原文摘要 · Abstract (English)
Time-series foundation models (TSFMs) have achieved strong univariate forecasting through large-scale pre-training, yet effectively extending this success to multivariate forecasting remains challenging. To address this, we propose DualWeaver, a novel framework that adapts univariate TSFMs (Uni-TSFMs) for multivariate forecasting by using a pair of learnable, structurally symmetric surrogate series. Generated by a shared auxiliary feature-fusion module that captures cross-variable dependencies, these surrogates are mapped to TSFM-compatible series via the forecasting objective. The symmetric structure enables parameter-free reconstruction of final predictions directly from the surrogates, without additional parametric decoding. A theoretically grounded regularization term is further introduced to enhance robustness against adaptation collapse. Extensive experiments on diverse real-world datasets show that DualWeaver outperforms state-of-the-art multivariate forecasters in both accuracy and stability. We release the code at https://github.com/li-jinpeng/DualWeaver.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。