Team, Then Trim: An Assembly-Line LLM Framework for High-Quality Tabular Data Generation

TL;DR

This paper introduces T$^2$, a novel framework using collaborative LLMs and a rigorous quality control pipeline to generate high-quality synthetic tabular data, addressing data scarcity and quality issues in ML applications.

Contribution

The paper presents a new assembly-line LLM framework with a three-stage QC pipeline for synthesizing superior tabular data, advancing data generation techniques.

Findings

T$^2$ outperforms existing methods in data quality metrics.

Empirical results show improved downstream model performance.

Framework effectively addresses class imbalance and bias issues.

Abstract

While tabular data is fundamental to many real-world machine learning (ML) applications, acquiring high-quality tabular data is usually labor-intensive and expensive. Limited by the scarcity of observations, tabular datasets often exhibit critical deficiencies, such as class imbalance, selection bias, and low fidelity. To address these challenges, building on recent advances in Large Language Models (LLMs), this paper introduces Team-then-Trim (T$^2$), a framework that synthesizes high-quality tabular data through a collaborative team of LLMs, followed by a rigorous three-stage plug-in data quality control (QC) pipeline. In T$^2$, tabular data generation is conceptualized as a manufacturing process: specialized LLMs, guided by domain knowledge, are tasked with generating different data components sequentially, and the resulting products, i.e., the synthetic data, are systematically evaluated across multiple dimensions of QC. Empirical results on both simulated and real-world datasets demonstrate that T$^2$ outperforms state-of-the-art methods in producing high-quality tabular data, highlighting its potential to support downstream models when direct data collection is practically infeasible.