Spatio-Temporal Trajectory Foundation Model - Recent Advances and Future Directions
It targets researchers in spatio-temporal data analytics, offering a review and future directions for TFMs, but is incremental as it synthesizes existing work rather than presenting new results.
This tutorial addresses the lack of systematic investigation into trajectory foundation models (TFMs), a subclass of spatio-temporal foundation models, by providing a comprehensive overview of recent advances, including a taxonomy and critical analysis of methodologies, strengths, and limitations.
Foundation models (FMs) have emerged as a powerful paradigm, enabling a diverse range of data analytics and knowledge discovery tasks across scientific fields. Inspired by the success of FMs, particularly large language models, researchers have recently begun to explore spatio-temporal foundation models (STFMs) to improve adaptability and generalization across a wide spectrum of spatio-temporal (ST) tasks. Despite rapid progress, a systematic investigation of trajectory foundation models (TFMs), a crucial subclass of STFMs, is largely lacking. This tutorial addresses this gap by offering a comprehensive overview of recent advances in TFMs, including a taxonomy of existing methodologies and a critical analysis of their strengths and limitations. In addition, the tutorial highlights open challenges and outlines promising research directions to advance spatio-temporal general intelligence through the development of robust, responsible, and transferable TFMs.