Dataset Inference for Data Provenance and Privacy Auditing in Tabular Foundation Models

Tabular foundation models (TFMs) are increasingly deployed through black-box APIs and trained on real-world tabular datasets. While private, proprietary, or otherwise unauthorized datasets may be incorporated into pre-training corpora, there are currently no dedicated methods for determining whether a given tabular dataset was used to train a TFM. As a solution, we introduce the first dataset inference method for TFMs, aiming to infer whether a suspect dataset was part of a model’s pre-training data. We systematically analyze a broad collection of candidate signals that can be observed from a black-box TFM via input manipulation and find that we can reliably infer dataset membership for several state-of-the-art TFMs trained on real tabular data, achieving up to 0.997 ROCAUC. We then study factors that influence dataset identification, including pre-training data composition, model capacity, and the use of real vs. synthetic data. Our results show that our dataset inference method is a practical auditing tool for detecting privacy leakage and the use of proprietary datasets in TFMs.