# Retrieval-Augmented Generation with Covariate Time Series

Source: [Retrieval-Augmented Generation with Covariate Time Series](<https://www.sota2.com/research/paper/69aa5f702d603e6de3e41813>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-26T05:00:04.121Z

Authors: Kenny Ye Liang, Zhongyi Pei, Huan Zhang, Yuhui Liu, Shaoxu Song, Jianmin Wang

Published: 2026-03-05T08:45:24.000Z

## Abstract

While RAG has greatly enhanced LLMs, extending this paradigm to Time-Series Foundation Models (TSFMs) remains a challenge. This is exemplified in the Predictive Maintenance of the Pressure Regulating and Shut-Off Valve (PRSOV), a high-stakes industrial scenario characterized by (1) data scarcity, (2) short transient sequences, and (3) covariate coupled dynamics. Unfortunately, existing time-series RAG approaches predominantly rely on generated static vector embeddings and learnable context augmenters, which may fail to distinguish similar regimes in such scarce, transient, and covariate coupled scenarios. To address these limitations, we propose RAG4CTS, a regime-aware, training-free RAG framework for Covariate Time-Series. Specifically, we construct a hierarchal time-series native knowledge base to enable lossless storage and physics-informed retrieval of raw historical regimes. We design a two-stage bi-weighted retrieval mechanism that aligns historical trends through point-wise and multivariate similarities. For context augmentation, we introduce an agent-driven strategy to dynamically optimize context in a self-supervised manner. Extensive experiments on PRSOV demonstrate that our framework significantly outperforms state-of-the-art baselines in prediction accuracy. The proposed system is deployed in Apache IoTDB within China Southern Airlines. Since deployment, our method has successfully identified one PRSOV fault in two months with zero false alarm.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Time-series forecasting on RPSOV B777L (MSE, MAE)](<https://www.sota2.com/research/sota/time-series-forecasting-on-rpsov-b777l-mse-mae.md>): Time Series Forecasting; RPSOV B777L; MSE: 0.058.

- [Time-series forecasting on RPSOV B777R](<https://www.sota2.com/research/sota/time-series-forecasting-on-rpsov-b777r.md>): Time Series Forecasting; RPSOV B777R; MSE: 0.095.

- [Time-series forecasting on RPSOV A320L (MSE, MAE)](<https://www.sota2.com/research/sota/time-series-forecasting-on-rpsov-a320l-mse-mae.md>): Time Series Forecasting; RPSOV A320L; MSE: 0.259.

- [Time-series forecasting on RPSOV A320R](<https://www.sota2.com/research/sota/time-series-forecasting-on-rpsov-a320r.md>): Time Series Forecasting; RPSOV A320R; MSE: 0.337.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.04951>)

[PDF](<https://arxiv.org/pdf/2603.04951.pdf>)
