Pretrained time series models, capable of zero-shot forecasting, have demonstrated significant potential in enhancing both the performance and accessibility of time series forecasting. However, existing pretrained models either do not support covariates or fail to incorporate them effectively. We introduce COSMIC, a zero-shot forecasting model that utilizes covariates via in-context learning. To address the challenge of data scarcity, we propose Informative Covariate Augmentation, which enables the training of COSMIC without requiring any datasets that include covariates. COSMIC achieves state-of-the-art performance in zero-shot forecasting, both with and without covariates. Our quantitative and qualitative analysis demonstrates that COSMIC effectively leverages covariates in zero-shot forecasting.
@article{arxiv.2506.03128,
title = {Zero-Shot Time Series Forecasting with Covariates via In-Context Learning},
author = {Andreas Auer and Raghul Parthipan and Pedro Mercado and Abdul Fatir Ansari and Lorenzo Stella and Bernie Wang and Michael Bohlke-Schneider and Syama Sundar Rangapuram},
journal= {arXiv preprint arXiv:2506.03128},
year = {2025}
}