TSPFN: A Temporal Tabular Foundation Model for Physiological Time Series Classification

arXiv:2608.31013 · cs.LG, cs.CV · Submitted 2026-08-31 · Read on arXiv

cs.LG, cs.CV

Submitted: 2026-08-31

Updated: 2026-08-31

Comments: Accepted at STACOM 2026 (MICCAI Workshop). 10 pages, 3 figures

Code: https://github.com/Jeremstym/TSPFN

License: http://creativecommons.org/licenses/by/4.0/

The gist: 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.

Terminology

Abstract

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

Sources

Related papers