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

2026-08-31

AI Summary

为了解决医疗时间序列分类中低至中等数据量下的泛化问题，本文提出TSPFN模型，通过整合结构化时间表示和位置嵌入来捕捉时间依赖性。

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

Problem

• physiological time series
• low-data regime
• temporal dependencies
• generalization

Innovation

• Temporal Tabular Foundation Model
• Structured Temporal Representations
• Positional Embeddings
• Cross-domain Generalization