发表机构
Université de Montréal; Mila – Quebec Artificial Intelligence Institute（蒙特利尔大学; 米拉-魁北克人工智能研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究探究表格基础模型（TFMs）是否掌握物理知识，通过在316个物理方程采样的数据集上对比4种TFMs与6种基准，发现TFMs性能更优，但先验无法表示无噪声机制与物理单位，尚不能作为物理模型使用。
AI 中文摘要
表格基础模型（Tabular Foundation Models, TFMs）学习以类似语言模型补全文本的方式补全表格，而表格可以说是大多数物理测量结果的呈现形式。它们在这个过程中是否学到了物理知识？TFMs在构造上是贝叶斯模型，因此问题在于它们的先验包含了什么。我们直接对其进行探测，在从316个物理方程中采样得到的域内和域外数据集上，评估了4种TFMs（TabPFN-3、TabICLv2、TabDPT和Real-TabPFN-2.5）与6种基准模型的性能。TFMs无论是开箱即用还是经过微调，都表现优于基准模型。但我们发现，它们的先验既无法表示无噪声机制，也无法表示物理单位，这就是它们能对物理现象进行插值，却还无法充当物理模型的原因。
英文摘要
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.
Comments5 pages (4 figures, 1 table). Submitted to the Representations for the Physical Sciences Workshop @ NeurIPS 2026
