aiqbench analysis
TSPFN addresses the critical challenge of generalizing machine learning models in low-data medical regimes by adapting the TabPFN architecture specifically for physiological time series. By integrating structured temporal representations and extensive pretraining on 140,000 real-world signals, this foundation model achieves superior cross-domain performance compared to both standard tabular baselines and specialized deep learning approaches. The work demonstrates a significant architectural shift required to handle spatio-temporal dependencies inherent in medical monitoring data without relying on conventional fine-tuning.
TSPFN: A Temporal Tabular Foundation Model for Physiological Time Series Classification
arXiv:2608.31013v1 Announce Type: new 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
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 this https URL
Accepted at STACOM 2026 (MICCAI Workshop). 10 pages, 3 figures
Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2608.31013 [cs.LG]
(or arXiv:2608.31013v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.31013 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 271 words
Related coverage
SemPSG: A Semantic Channel-Aware Foundation Model for Polysomnography Analysis
arXiv · Sep 30, 2026
Towards Unified and Data-Efficient Prognostics and Health Management with Tabular Foundation Models
arXiv · Jun 6, 2026
7 Steps to Mastering Time Series Analysis with Python
KDnuggets · Jun 4, 2026
In-Context Time Series Classification with Random Convolutional Features
arXiv · Jul 22, 2026