Robust Detection of Synthetic Tabular Data under Schema Variability

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kindji, G. Charbel N., Fromont, Elisa, Rojas-Barahona, Lina Maria, Urvoy, Tanguy
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918225078386688
author Kindji, G. Charbel N.
Fromont, Elisa
Rojas-Barahona, Lina Maria
Urvoy, Tanguy
author_facet Kindji, G. Charbel N.
Fromont, Elisa
Rojas-Barahona, Lina Maria
Urvoy, Tanguy
contents The rise of powerful generative models has sparked concerns over data authenticity. While detection methods have been extensively developed for images and text, the case of tabular data, despite its ubiquity, has been largely overlooked. Yet, detecting synthetic tabular data is especially challenging due to its heterogeneous structure and unseen formats at test time. We address the underexplored task of detecting synthetic tabular data ''in the wild'', i.e. when the detector is deployed on tables with variable and previously unseen schemas. We introduce a novel datum-wise transformer architecture that significantly outperforms the only previously published baseline, improving both AUC and accuracy by 7 points. By incorporating a table-adaptation component, our model gains an additional 7 accuracy points, demonstrating enhanced robustness. This work provides the first strong evidence that detecting synthetic tabular data in real-world conditions is feasible, and demonstrates substantial improvements over previous approaches. Following acceptance of the paper, we are finalizing the administrative and licensing procedures necessary for releasing the source code. This extended version will be updated as soon as the release is complete.
format Preprint
id arxiv_https___arxiv_org_abs_2509_00092
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Robust Detection of Synthetic Tabular Data under Schema Variability
Kindji, G. Charbel N.
Fromont, Elisa
Rojas-Barahona, Lina Maria
Urvoy, Tanguy
Machine Learning
Artificial Intelligence
Databases
The rise of powerful generative models has sparked concerns over data authenticity. While detection methods have been extensively developed for images and text, the case of tabular data, despite its ubiquity, has been largely overlooked. Yet, detecting synthetic tabular data is especially challenging due to its heterogeneous structure and unseen formats at test time. We address the underexplored task of detecting synthetic tabular data ''in the wild'', i.e. when the detector is deployed on tables with variable and previously unseen schemas. We introduce a novel datum-wise transformer architecture that significantly outperforms the only previously published baseline, improving both AUC and accuracy by 7 points. By incorporating a table-adaptation component, our model gains an additional 7 accuracy points, demonstrating enhanced robustness. This work provides the first strong evidence that detecting synthetic tabular data in real-world conditions is feasible, and demonstrates substantial improvements over previous approaches. Following acceptance of the paper, we are finalizing the administrative and licensing procedures necessary for releasing the source code. This extended version will be updated as soon as the release is complete.
title Robust Detection of Synthetic Tabular Data under Schema Variability
topic Machine Learning
Artificial Intelligence
Databases
url https://arxiv.org/abs/2509.00092