cs.LGNov 13, 2026

T2ST: Data-Efficient Spatio-Temporal Foundation Models via Temporal Data and Knowledge Transfer

Authors: Yue Xu, Kaiyun Wei, Wenying Duan, Zichen Xu, Shuai Ma, Xiaoxi He

Organizations: Faculty of Information Science and Computing, University of Macau, Macau, China · School of Artificial Intelligence, Nanchang University, Nanchang, China · School of Computer Science Intelligence, Nanchang University, Nanchang, China · School of Computer Science and Engineering, Beihang University, Beijing, China

Abstract

Spatio-temporal foundation models (STFMs) require large-scale training corpora, yet synchronized observations and spatial metadata are costly to acquire and curate, limiting the availability of suitable data. By contrast, time-series data are abundant and easier to collect. Since observations at each location form a time series, this motivates a natural question: can abundant time-series data enhance STFMs when spatio-temporal observations are scarce? To bridge this gap, we propose T2ST, a data-efficient training framework that augments limited spatio-temporal observations with selected auxiliary time series and transfers knowledge from a frozen time-series foundation model. Our key insight is that temporal pattern learning can benefit from abundant external time series, while spatial dependency learning remains grounded in real spatio-temporal observations. Guided by this insight, T2ST retrieves matched sequences to reinforce sparsely observed temporal patterns and selects reliable complementary sequences to broaden temporal coverage. Temporal relation distillation aligns relationships among historical patches, while multi-scale prediction distillation transfers forecasting behavior at fine and coarse temporal resolutions, enabling knowledge transfer across heterogeneous architectures. Using only 10% of the original spatio-temporal data, T2ST enables STFMs to achieve overall forecasting performance competitive with or superior to their full-data counterparts across six long-term and twelve short-term forecasting datasets. These results establish a new paradigm for data-efficient training of spatio-temporal foundation models. Code is available at https://anonymous.4open.science/r/T2ST-65F3/.

Figures & tables

Explore similar work

CardsList
  1. Decouple to Reuse: Fine-Grained Knowledge Augmentation for Time Series Foundation Models

    Nov 4, 2026Yuqian Wu, Qingxiang Liu, Yuxin Guo +6Time-Series Foundation ModelsRetrieval-Augmented Time Series Forecasting

  2. TimeNet: An Extensible Unified Data Infrastructure for Next-Generation Temporal Foundation Models

    Oct 3, 2026Martin Maritsch, Timo Stoffregen, Thomas Kaar +36Time-Series Foundation Models

  3. Kairos: Toward Adaptive and Parameter-Efficient Time Series Foundation Models

    Sep 30, 2025Kun Feng, Shaocheng Lan, Yuchen Fang +6Rotary Positional EmbeddingsZero-Shot Time Series Forecasting