A Survey on Tabular Data Generation: Utility, Alignment, Fidelity, Privacy, and Beyond

2025-03-07

AI Summary

This paper addresses four practical requirements in tabular data generation: utility, domain-knowledge alignment, statistical fidelity, and privacy preservation. To this end, it proposes the first four-dimensional, requirement-driven classification framework, systematically surveying over 120 works and establishing explicit mappings between model families (e.g., GANs, VAEs, diffusion models, normalizing flows, and LLM-based approaches) and targeted requirements. Methodologically, it integrates deep generative modeling, multi-dimensional quantitative evaluation, domain-knowledge injection, and differential privacy mechanisms. Critically, it identifies fundamental limitations in existing evaluation paradigms and prescribes concrete improvements. The work delivers a structured, taxonomy-guided survey and a practical, requirement-aware model selection and evaluation guideline—advancing tabular generation research toward demand-driven, verifiable, trustworthy, and controllable outcomes.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Generative modelling has become the standard approach for synthesising tabular data. However, different use cases demand synthetic data to comply with different requirements to be useful in practice. In this survey, we review deep generative modelling approaches for tabular data from the perspective of four types of requirements: utility of the synthetic data, alignment of the synthetic data with domain-specific knowledge, statistical fidelity of the synthetic data distribution compared to the real data distribution, and privacy-preserving capabilities. We group the approaches along two levels of granularity: (i) based on the primary type of requirements they address and (ii) according to the underlying model they utilise. Additionally, we summarise the appropriate evaluation methods for each requirement and the specific characteristics of each model type. Finally, we discuss future directions for the field, along with opportunities to improve the current evaluation methods. Overall, this survey can be seen as a user guide to tabular data generation: helping readers navigate available models and evaluation methods to find those best suited to their needs.

Problem

• Review deep generative models for tabular data synthesis.
• Evaluate synthetic data utility, alignment, fidelity, and privacy.
• Guide users in selecting models and evaluation methods.

Innovation

• Deep generative models for tabular data synthesis.
• Focus on utility, alignment, fidelity, and privacy.
• Evaluation methods tailored to specific requirements.