Page Archive

TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster

Published in arXiv preprint, 2025

This paper proposes TS-RAG, a retrieval-augmented generation based time series forecasting framework that enhances the generalization capability and interpretability of time series foundation models.

Recommended citation: Kanghui Ning, Zijie Pan, Yu Liu, Yushan Jiang, James Y. Zhang, Kashif Rasul, Anderson Schneider, Lintao Ma, Yuriy Nevmyvaka, Dongjin Song. (2025). "TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster." arXiv preprint arXiv:2503.07649.
Download Paper