Privately generating tabular data using language models

2023-06-07 · Alexandre Sablayrolles, Yue Wang, Brian Karrer

Privately generating synthetic data from a table is an important brick of a privacy-first world. We propose and investigate a simple approach of treating each row in a table as a sentence and training a language model with differential privacy. We show this approach obtains competitive results in modelling tabular data across multiple datasets, even at small scales that favor alternative methods based on marginal distributions.

📄 PDF Abstract BibTeX arXiv:2306.04803

Code (1)

facebookresearch/synlm 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingSentence

Differentially Private Synthetic Data via APIs 4: Tabular Data

2026-06-06 · Toan Tran, Arturs Backurs, Zinan Lin, Victor Reis 외 arxiv

This paper investigates the problem of generating synthetic tabular data with differential privacy (DP) guarantees, enabling data sharing in sensitive domains. Despite extensive study, state-of-the-art methods often focu…

Tabular Data Generation

Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Data

2026-02-02 · Lucas Rosenblatt, Peihan Liu, Ryan McKenna, Natalia Ponomareva arxiv

Research on differentially private synthetic tabular data has largely focused on independent and identically distributed rows where each record corresponds to a unique individual. This perspective neglects the temporal c…

TabPATE: Differentially Private Tabular In-Context Learning Without Public Data

2026-06-30 · Dariush Wahdany, Matthew Jagielski, Jesse C. Cresswell, Adam Dziedzic 외 arxiv

Tabular foundation models enable accurate in-context learning (ICL) from small labeled datasets, but the private records placed in context can leak through model predictions. We first show that even basic membership infe…

EPIC: Effective Prompting for Imbalanced-Class Data Synthesis in Tabular Data Classification via Large Language Models

2024-04-15 · Jinhee Kim, Taesung Kim, Jaegul Choo

Large language models (LLMs) have demonstrated remarkable in-context learning capabilities across diverse applications. In this work, we explore the effectiveness of LLMs for generating realistic synthetic tabular data, …

In-Context Learning Synthetic Data Generation Tabular Data Generation

DP-2Stage: Adapting Language Models as Differentially Private Tabular Data Generators

2024-12-03 · Tejumade Afonja, Hui-Po Wang, Raouf Kerkouche, Mario Fritz

Generating tabular data under differential privacy (DP) protection ensures theoretical privacy guarantees but poses challenges for training machine learning models, primarily due to the need to capture complex structures…