SEED: Enhancing Text-to-SQL Performance and Practical Usability Through Automatic Evidence Generation

TL;DR

SEED automatically generates evidence from database schemas to enhance text-to-SQL performance and usability, reducing reliance on human-provided evidence and improving model robustness in real-world applications.

Contribution

The paper introduces SEED, a novel automatic evidence generation method that improves text-to-SQL accuracy and practicality without requiring human-annotated evidence.

Findings

SEED significantly improves SQL accuracy in no-evidence scenarios.

SEED can outperform models with human-provided evidence in some cases.

Enhances model robustness and adaptability for real-world deployment.

Abstract

Text-to-SQL enables non-experts to retrieve data from databases by converting natural language queries into SQL. However, state-of-the-art text-to-SQL studies rely on the BIRD dataset, which assumes that evidence is provided along with questions. Although BIRD facilitates research advancements, it assumes that users have expertise and domain knowledge, contradicting the fundamental goal of text-to-SQL. In addition, human-generated evidence in BIRD contains defects, including missing or erroneous evidence, which affects model performance. To address this issue, we propose SEED (System for Evidence Extraction and Domain knowledge generation), an approach that automatically generates evidence to improve performance and practical usability in real-world scenarios. SEED systematically analyzes database schema, description files, and values to extract relevant information. We evaluated SEED on BIRD and Spider, demonstrating that it significantly improves SQL generation accuracy in the no-evidence scenario, and in some cases, even outperforms the setting where BIRD evidence is provided. Our results highlight that SEED-generated evidence not only bridges the gap between research and real-world deployment but also improves the adaptability and robustness of text-to-SQL models. Our code is available at https://github.com/felix01189/SEED