DCMM-SQL: Automated Data-Centric Pipeline and Multi-Model Collaboration Training for Text-to-SQL Model

TL;DR

This paper introduces an automated data-centric pipeline and multi-model collaboration training approach for text-to-SQL tasks, significantly improving accuracy by repairing, augmenting data, and leveraging multiple models.

Contribution

It proposes a novel fully automated data-centric pipeline with adaptive data repair and error data augmentation, along with a multi-model collaboration training schema for text-to-SQL.

Findings

Achieved first place in lightweight text-to-SQL models (within 70B).

Demonstrated effectiveness of data repair and augmentation strategies.

Validated multi-model collaboration improves task accuracy.

Abstract

Text-to-SQL tasks have gained attractive improvements since the release of ChatGPT. Among them, agent-based frameworks have been widely used in this field. However, the impact of data-centric strategies on text-to-SQL tasks has rarely been explored. In this paper, we systemically design a fully automated data-centric pipeline for text-to-SQL tasks, including \emph{adaptive data repair}, which can automatically find and fix errors in the training dataset; and \emph{error data augmentation}, where we specifically diffuse and enhance erroneous data predicted by the initially trained models. Meanwhile, we propose a Multi-Model collaboration training schema, aiming to train multiple models with different augmented data, enabling them to possess distinct capabilities and work together to complement each other, because it has been found that the capability of a single fine-tuned model is very limited. Furthermore, we utilize an ensemble strategy to integrate the capabilities of multiple models to solve a multiple-choice question, aiming to further improve the accuracy of text-to-SQL tasks. The experiment results and ablation study have demonstrated the effectiveness of data-centric pipeline and Multi-Model(MM) interactive iterative strategies, achieving first place in lightweight text-to-SQL models (within 70B).