Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit

2026-09-02

AI Summary

研究通过对比四个表格基础模型与六个基线在316个物理方程数据集上的表现，探讨这些模型是否学到了物理知识，发现它们虽然表现出色但不能代表无噪声机制或物理单位。

Abstract

Tabular foundation models (TFMs) learn to fill in tables the way language models fill in text, and tables are arguably the format in which most physical measurement arrives. Did they learn any physics in the process? They are Bayesian by construction, so the question is what their prior contains. We probe it directly, evaluating four of them (TabPFN-3, TabICLv2, TabDPT and Real-TabPFN-2.5) against six baselines on datasets sampled from 316 physical equations, in and out of domain. TFMs dominate, out of the box and after tuning. But we show that their prior can represent neither a noiseless mechanism nor physical units, which is why they interpolate physics without yet being able to act as physical models.

Problem

• Tabular Foundation Models
• Physics
• Prior
• Units
• Mechanism

Innovation

• Tabular Foundation Models
• Physics Learning
• Bayesian Priors
• Physical Units
• Interpolation vs. Physical Modeling