[2506.22039] UniCA: Unified Covariate Adaptation for Time Series Foundation Model

Computer Science > Machine Learning arXiv:2506.22039 (cs) [Submitted on 27 Jun 2025 (v1), last revised 24 Mar 2026 (this version, v2)] Title:UniCA: Unified Covariate Adaptation for Time Series Foundation Model Authors:Lu Han, Yu Liu, Lan Li, Qiwen Deng, Jian Jiang, Yinbo Sun, Zhe Yu, Binfeng Wang, Xingyu Lu, Lintao Ma, Han-Jia Ye, De-Chuan Zhan View a PDF of the paper titled UniCA: Unified Covariate Adaptation for Time Series Foundation Model, by Lu Han and 11 other authors View PDF HTML (experimental) Abstract:Time Series Foundation Models (TSFMs) have achieved remarkable success through large-scale pretraining. However, their design primarily targets real-valued series, limiting their ability to handle general forecasting tasks involving diverse and often heterogeneous covariates -- such as categorical variables and multimodal data (e.g., images, text) -- which are typically task-specific and difficult to leverage during pretraining. To address this gap, we propose Unified Covariate Adaptation (UniCA), a framework to bridge TSFMs with general covariate-aware forecasting. UniCA first performs covariate homogenization to transform heterogeneous covariates into high-level homogeneous series representations and then fuses them via a unified attention-based fusion mechanism. UniCA is compatible and universal for adaptation with both homogeneous and heterogeneous covariates, incorporating extra covariate information while preserving the generalization ability of this http URL ...