aiqbench analysis
SARAF addresses the brittleness of standard retrieval-augmented forecasting by dynamically balancing relevance and diversity to handle non-stationary data regimes. The framework automatically adjusts its selection strategy based on dataset-level stationarity, offering improved robustness over strong baselines in challenging real-world scenarios. This approach provides a practical solution for time series models that must adapt to shifting historical patterns without relying solely on parametric assumptions.
Stationarity-Aware Retrieval-Augmented Time Series Forecasting
arXiv:2606.04135v1 Announce Type: new 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.
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: this https URL.
Machine Learning (cs.LG)
arXiv:2606.04135 [cs.LG]
(or arXiv:2606.04135v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.04135 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 262 words
Related coverage
Which Histories Matter for Time Series Forecasting? Learning Predictive Relevance with Future Supervision
arXiv · Aug 25, 2026
Cracking the Data Science Case Study Interview
Analytics Vidhya · Jul 26, 2026
Semantics-Enhanced Retrieval-Augmented Time Series Forecasting
arXiv · Jun 16, 2026
TFGformer: Multivariate Time Series Forecasting via Time-Frequency Graph Learning and Covariate Fusion
arXiv · Aug 3, 2026