Understanding the Surprising Generalization Properties of Tabular Foundation Models
Tabular Foundation Models (TFMs) increasingly rely on in-context learning, where a model receives labelled examples at inference time and predicts labels for new inputs without updating its weights. Existing TFMs are typically trained on either massive synthetic corpora or very large collections of real datasets. In contrast, we show that surprisingly strong transfer can emerge from self-supervised pre-training on just a single real table. In this setting, we also find that tables tend to be either broadly useful or broadly poor regardless of downstream prediction task, and that the strongest
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 62%Google's TabFM skips per-dataset training and still predicts on tables it's never seen →
- PossiblePossibly related (embedding) · 56%TabFM Studio: point-and-click predictions on spreadsheets with tabular foundation models, fully local [P] →
- LinkedLinked via arxiv author · 85%Nour Shaheen →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”
- LinkedLinked via arxiv author · 85%Junwei Ma →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”
- LinkedLinked via arxiv author · 85%Alex Labach →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”
- LinkedLinked via arxiv author · 85%Frank Hutter →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”
- LinkedLinked via arxiv author · 85%Valentin Thomas →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”
- LinkedLinked via arxiv author · 85%Anthony L. Caterini →
“Understanding the Surprising Generalization Properties of Tabular Foundation Models”