# TAB-AUDIT: Detecting AI-Fabricated Scientific Tables via Multi-View Likelihood Mismatch

Source: [TAB-AUDIT: Detecting AI-Fabricated Scientific Tables via Multi-View Likelihood Mismatch](<https://www.sota2.com/research/paper/69c0c8dc0636e5bf4afc3997>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-24T05:00:03.763Z

Authors: Shuo Huang, Yan Pen, Lizhen Qu

Published: 2026-03-20T07:33:54.000Z

## Abstract

AI-generated fabricated scientific manuscripts raise growing concerns with large-scale breaches of academic integrity. In this work, we present the first systematic study on detecting AI-generated fabricated scientific tables in empirical NLP papers, as information in tables serve as critical evidence for claims. We construct FabTab, the first benchmark dataset of fabricated manuscripts with tables, comprising 1,173 AI-generated papers and 1,215 human-authored ones in empirical NLP. Through a comprehensive analysis, we identify systematic differences between fabricated and real tables and operationalize them into a set of discriminative features within the TAB-AUDIT framework. The key feature, within-table mismatch, captures the perplexity gap between a table's skeleton and its numerical content. Experimental results show that RandomForest built on these features significantly outperform prior state-of-the-art methods, achieving 0.987 AUROC in-domain and 0.883 AUROC out-of-domain. Our findings highlight experimental tables as a critical forensic signal for detecting AI-generated scientific fraud and provide a new benchmark for future research.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [AI-generated paper detection on GPT Clean Holdout 5.2 (test)](<https://www.sota2.com/research/sota/ai-generated-paper-detection-on-gpt-clean-holdout-5-2-test.md>): AI-generated paper detection; GPT Clean Holdout 5.2 (test); AUROC: 0.8857.

- [AI-generated paper detection on Prompt-only baseline (test)](<https://www.sota2.com/research/sota/ai-generated-paper-detection-on-prompt-only-baseline-test.md>): AI-generated paper detection; Prompt-only baseline (test); AUROC: 0.973.

- [AI-generated table detection on GPT 5.2 (External Holdout)](<https://www.sota2.com/research/sota/ai-generated-table-detection-on-gpt-5-2-external-holdout.md>): AI-generated table detection; GPT 5.2 (External Holdout); AUROC: 88.1.

- [Fabricated Table Detection on FABTAB GPT-4o In-Domain (test)](<https://www.sota2.com/research/sota/fabricated-table-detection-on-fabtab-gpt-4o-in-domain-test.md>): Fabricated Table Detection; FABTAB GPT-4o In-Domain (test); AUROC: 0.902.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.19712>)

[PDF](<https://arxiv.org/pdf/2603.19712.pdf>)
