Track-SQL: Enhancing Generative Language Models with Dual-Extractive Modules for Schema and Context Tracking in Multi-turn Text-to-SQL

TL;DR

Track-SQL introduces dual-extractive modules to improve multi-turn Text-to-SQL performance by effectively tracking schema and context changes, achieving state-of-the-art results.

Contribution

The paper presents a novel framework with semantic-enhanced schema and context extractors to address multi-turn Text-to-SQL challenges, outperforming existing models.

Findings

Achieves state-of-the-art performance on SparC and CoSQL datasets.

Improves execution accuracy by 7.1% and 9.55% respectively.

Significantly enhances multi-turn interaction handling.

Abstract

Generative language models have shown significant potential in single-turn Text-to-SQL. However, their performance does not extend equivalently to multi-turn Text-to-SQL. This is primarily due to generative language models' inadequacy in handling the complexities of context information and dynamic schema linking in multi-turn interactions. In this paper, we propose a framework named Track-SQL, which enhances generative language models with dual-extractive modules designed to track schema and contextual changes in multi-turn Text-to-SQL. Specifically, Track-SQL incorporates a \emph{Semantic-enhanced Schema Extractor} and a \emph{Schema-aware Context Extractor}. Experimental results demonstrate that Track-SQL achieves state-of-the-art performance on the SparC and CoSQL datasets. Furthermore, detailed ablation studies reveal that Track-SQL significantly improves execution accuracy in multi-turn interactions by 7.1\% and 9.55\% on these datasets, respectively. Our implementation will be open-sourced at https://github.com/DMIRLAB-Group/Track-SQL.