Search papers, labs, and topics across Lattice.
This paper introduces TSPFN, a novel foundation model specifically designed for physiological time-series classification by enhancing the architecture of TabPFN to effectively capture temporal dependencies. By integrating structured temporal representations and positional embeddings, TSPFN is pretrained on a substantial dataset of 140,000 physiological time series, enabling it to generalize across various medical domains. Experimental results reveal that TSPFN not only surpasses traditional tabular models and TabPFN but also outperforms specialized deep time-series models in cross-domain generalization tasks.
TSPFN outperforms both traditional tabular models and specialized deep learning approaches in classifying physiological time series, showcasing its superior generalization capabilities.
Designing models that generalize effectively in low- to medium-data regimes remains a primary challenge in medical machine learning, particularly for physiological time-series classification. While tabular foundation models such as TabPFN offer an attractive alternative to conventional fine-tuning through in-context learning, they are not designed to capture the temporal dependencies inherent to physiological signals. ~In this paper, we introduce TSPFN, a foundation model that redesigns TabPFN's architecture for time series data. TSPFN integrates structured temporal representations and positional embeddings to capture intra-sample temporal and channel dependencies. To fully leverage its spatio-temporal design, the model is pretrained on 140,000 real-world physiological time series across multiple medical domains. This yields a unified, generalizable framework capable of learning the specificities of medical time series. Experiments across diverse physiological benchmarks demonstrate that TSPFN consistently outperforms standard tabular baselines and TabPFN, and achieves superior cross-domain generalization compared to specialized deep time-series models. All our experiments, ablation studies, and pre-processing scheme are publicly available at https://github.com/Jeremstym/TSPFN
