Datum-wise Transformer for Synthetic Tabular Data Detection in the Wild

2025-04-10

AI Summary

Detecting synthetic tabular data in real-world scenarios remains challenging due to unknown table structures and poor cross-table generalization. Method: This paper proposes a datum-wise Transformer architecture that models each row independently, eliminating assumptions about column count or type. It incorporates column-aware embedding, dynamic masked attention, and adversarial domain alignment to enable zero-shot structural transfer and robust detection across heterogeneous distributions. Contribution/Results: This work is the first to introduce the datum-wise paradigm to synthetic tabular data detection, supporting arbitrary, previously unseen table schemas. Evaluated on a multi-source, heterogeneous benchmark, our method achieves an 8.2% AUC improvement over state-of-the-art approaches. The framework delivers scalable, highly generalizable synthetic-data attribution—critical for industrial and governmental applications requiring reliable provenance tracing of tabular content.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The growing power of generative models raises major concerns about the authenticity of published content. To address this problem, several synthetic content detection methods have been proposed for uniformly structured media such as image or text. However, little work has been done on the detection of synthetic tabular data, despite its importance in industry and government. This form of data is complex to handle due to the diversity of its structures: the number and types of the columns may vary wildly from one table to another. We tackle the tough problem of detecting synthetic tabular data ''in the wild'', i.e. when the model is deployed on table structures it has never seen before. We introduce a novel datum-wise transformer architecture and show that it outperforms existing models. Furthermore, we investigate the application of domain adaptation techniques to enhance the effectiveness of our model, thereby providing a more robust data-forgery detection solution.

Problem

• Detecting synthetic tabular data in diverse real-world structures
• Addressing lack of methods for non-uniform tabular data detection
• Improving robustness with domain adaptation for data-forgery detection

Innovation

• Datum-wise transformer for diverse tabular data
• Domain adaptation enhances detection robustness
• Outperforms existing synthetic data detection models