用动态检索提升时间序列预测，解决异源数据不一致问题
TFGformer: Multivariate Time Series Forecasting via Time-Frequency Graph Learning and Covariate Fusion
- 引入形状感知记忆与逆归一化，实现跨量级数据的形状匹配检索
- 通过未来一致性对比学习，区分历史相似但未来不同的干扰样本
- 融合检索结果到模型表示层，显著提升长周期预测精度
来自异构物联网传感器的大规模多变量时间序列，对资源调度和预测性维护的长期准确预测提出需求。尽管近期时间序列基础模型具备强泛化能力，但依赖静态参数知识，在推理阶段缺乏对外部历史模式的动态访问。检索增强生成（RAG）提供潜在解决方案，但其在时间序列预测中的应用受限于异源数据间的幅度差异，以及历史相似性与未来一致性之间的不匹配。我们提出CrossRAG，一个融合形状感知记忆（SAM）与RevIN归一化的检索增强预测框架，通过未来一致性对比学习（FCC）区分具有相似历史但未来分歧的负样本，并采用跨注意力时序融合（CATF）在表示层面融合检索到的历史-未来参考对。在七个公开基准上的实验表明，CrossRAG在多数场景下优于仅使用参数的基线及现有检索增强方法。
原文摘要 · Abstract (English)
Large-scale multivariate time series from heterogeneous IoT sensors demand accurate long-term forecasting for resource scheduling and predictive maintenance. While recent time series foundation models exhibit strong generalization, they rely on static parametric knowledge and lack dynamic access to external historical patterns during inference. Retrieval-Augmented Generation (RAG) offers a potential remedy, yet its application to time series forecasting is challenged by magnitude variations across heterogeneous sources and the mismatch between historical similarity and future consistency. We propose CrossRAG, a retrieval-augmented forecasting framework that integrates Shape-Aware Memory (SAM) with RevIN normalization for magnitude-robust shape-level retrieval, Future-Consistent Contrastive (FCC) learning to distinguish informative references from hard negatives with similar history but divergent futures, and Cross-Attention Temporal Fusion (CATF) to fuse retrieved historical--future reference pairs into the backbone's representations at the representation level. Experiments on seven public benchmarks show that CrossRAG consistently outperforms both parametric-only baselines and existing retrieval-augmented forecasting methods.
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。