# LLM-as-a-Discriminator: When Synthetic Tables Still Look Real

arXiv: https://arxiv.org/abs/2606.09865 · HTML: https://mlsanity.com/paper/2606.09865/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 26 references that are in the map.
- First posted on arXiv 1 Jun 2026.

## Builds on (most cited first, 14 of 14)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [Modeling Tabular data using Conditional GAN](https://mlsanity.com/paper/1907.00503/) (2019)
- [TabDDPM: Modelling Tabular Data with Diffusion Models](https://mlsanity.com/paper/2209.15421/) (2022)
- [Language Models are Realistic Tabular Data Generators](https://mlsanity.com/paper/2210.06280/) (2022)
- [Large Language Models(LLMs) on Tabular Data: Prediction, Generation, and Understanding -- A Survey](https://mlsanity.com/paper/2402.17944/) (2024)
- [Synthetic Data -- Anonymisation Groundhog Day](https://mlsanity.com/paper/2011.07018/) (2020)
- [Are Attribute Inference Attacks Just Imputation?](https://mlsanity.com/paper/2209.01292/) (2022)
- [Mind the Confidence Gap: Overconfidence, Calibration, and Distractor Effects in Large Language Models](https://mlsanity.com/paper/2502.11028/) (2025)
- [Judge's Verdict: A Comprehensive Analysis of LLM Judge Capability Through Human Agreement](https://mlsanity.com/paper/2510.09738/) (2025)
- [Training an LLM-as-a-Judge Model: Pipeline, Insights, and Practical Lessons](https://mlsanity.com/paper/2502.02988/) (2025)
- [Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata](https://mlsanity.com/paper/2207.03339/) (2022)

## Abstract (arXiv)

Privacy and data sharing are often in tension. Many organizations use synthetic data to reduce privacy risk and still share useful data. For tabular data, auditing privacy remains hard. In many cases, even humans cannot easily tell if a table is real or synthetic. In this paper, we propose a method based on LLM discrimination. We ask an LLM to classify each table sample as REAL or SYNTHETIC. We test two settings: C1 with table only, and C2 with table plus distributional metadata. We use LLaMA as an open model and Gemini as a reference model. In our experiments, we run three synthesis models, CTGAN, TVAE, and Gaussian Copula, on two public datasets, UCI Adult and ACS Census. We collect 451 valid trials. Our results show clear differences between models. On Adult, LLaMA reaches DRS=0% in reported cells, while Gemini reaches DRS=100% for CTGAN and TVAE. On Census, LLaMA predicts SYNTHETIC for most samples, while Gemini stays high in C1 but drops for CTGAN and TVAE in C2. We also compare with a classifier two-sample test (C2ST) and record linkage as distributional baselines, and with a human pilot of 2 annotators and 240 trials. Our results show that LLM discrimination is a practical privacy audit signal when model choice, per provider reporting, and data encoding are handled with care. For reproducibility, code and experiment scripts are available at https://github.com/SlokomManel/LLM-as-a-Discriminator.