Cross-city Time Series Forecasting with Retrieval-Augmented Large Language Models.

Yue Jiang, Chenxi Liu, Yile Chen, Qin Chao, Shuai Liu, Long Cheng, Gao Cong.

“Cross-city Time Series Forecasting with Retrieval-Augmented Large Language Models” proposes a selective knowledge transfer framework for urban time-series forecasting that retrieves only semantically and temporally relevant source subsequences using a patch-based encoder and integrates them with a reasoning-capable model to improve predictions in data-scarce cities.