TFGformer: Multivariate Time Series Forecasting via Time-Frequency Graph Learning and Covariate Fusion
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…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
