Channel-wise Retrieval for Multivariate Time Series Forecasting

2026-04-07

AI Summary

This work addresses the challenges of modeling long-range dependencies in multivariate time series forecasting and the neglect of inter-variable heterogeneity in existing retrieval-augmented methods. To this end, the authors propose CRAFT, a novel framework that introduces, for the first time, a channel-wise retrieval mechanism. CRAFT employs a two-stage efficient retrieval pipeline: in the time domain, it prunes candidate segments using a sparse relational graph, and in the frequency domain, it ranks reference segments based on spectral similarity. This design enables precise capture of each variable’s unique periodicity and spectral characteristics. Extensive experiments demonstrate that CRAFT significantly outperforms state-of-the-art methods across seven public benchmarks, achieving markedly improved prediction accuracy while maintaining practical inference efficiency.

Abstract

Multivariate time series forecasting often struggles to capture long-range dependencies due to fixed lookback windows. Retrieval-augmented forecasting addresses this by retrieving historical segments from memory, but existing approaches rely on a channel-agnostic strategy that applies the same references to all variables. This neglects inter-variable heterogeneity, where different channels exhibit distinct periodicities and spectral profiles. We propose CRAFT (Channel-wise retrieval-augmented forecasting), a novel framework that performs retrieval independently for each channel. To ensure efficiency, CRAFT adopts a two-stage pipeline: a sparse relation graph constructed in the time domain prunes irrelevant candidates, and spectral similarity in the frequency domain ranks references, emphasizing dominant periodic components while suppressing noise. Experiments on seven public benchmarks demonstrate that CRAFT outperforms state-of-the-art forecasting baselines, achieving superior accuracy with practical inference efficiency.

Problem

• multivariate time series forecasting
• channel heterogeneity
• long-range dependencies
• retrieval-augmented forecasting
• spectral profiles

Innovation

• channel-wise retrieval
• multivariate time series forecasting
• spectral similarity
• retrieval-augmented forecasting
• sparse relation graph