Stationarity-Aware Retrieval-Augmented Time Series Forecasting
2026-06-02 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address challenges in time series forecasting caused by changing patterns over time, which make it hard to rely only on past data. They propose a method called SARAF that improves predictions by carefully selecting diverse and relevant past examples based on how stable a dataset is. Their approach mixes these past segments in a smart way to better handle sudden changes in the data. Tests on real-world data show SARAF performs well, especially when dealing with non-stationary time series where patterns shift frequently.
Time Series ForecastingNon-StationarityRegime ShiftsRetrieval-Augmented Generation (RAG)Temporal SimilarityDiversity-aware SelectionStationarityInferenceRobustness
Authors
Shiqiao Zhou, Holger Schöner, Zipeng Wu, Edouard Fouché, IAG Wilson, Shuo Wang
Abstract
Time series forecasting relies on historical patterns, but real-world series often exhibit non-stationarity and regime shifts that challenge fully parametric forecasters. Inspired by Retrieval-Augmented Generation (RAG), recent work augments forecasters by retrieving relevant historical segments and using them as external evidence at inference time. However, due to the intrinsic non-stationarity of real-world time series, a highly similar past segment does not necessarily imply a similar future, rendering similarity-only retrieval brittle and prone to redundancy. We propose Stationarity-Aware Retrieval-Augmented Time Series Forecasting (SARAF), a framework that adaptively balances relevance and diversity in retrieval. SARAF first forms a candidate pool via temporal similarity with time-aligned enhancement, then applies a diversity-aware selection strategy to cover heterogeneous historical regimes, with the diversification strength automatically modulated by dataset-level stationarity. Moreover, SARAF uses stationarity-aware aggregation to fuse the retrieved futures. Extensive experiments on eight real-world datasets show that SARAF achieves competitive forecasting performance and improves average accuracy and robustness over strong baselines, with particularly clear benefits under challenging non-stationary settings. Code: https://github.com/ShiqiaoZhou/SARAF.