Knowledge Base Construction for Knowledge-Augmented Text-to-SQL
This addresses the challenge of domain-specific and diverse queries in text-to-SQL systems, enabling more accurate SQL generation for users interacting with databases, though it is incremental as it builds on existing knowledge-augmented approaches.
The paper tackles the problem of limited parametric knowledge in Large Language Models for text-to-SQL tasks by constructing a comprehensive knowledge base from questions, database schemas, and relevant knowledge, which improves accuracy and generalizes to unseen databases. It demonstrates substantial performance gains over baselines on multiple datasets in both overlapping and non-overlapping database scenarios.
Text-to-SQL aims to translate natural language queries into SQL statements, which is practical as it enables anyone to easily retrieve the desired information from databases. Recently, many existing approaches tackle this problem with Large Language Models (LLMs), leveraging their strong capability in understanding user queries and generating corresponding SQL code. Yet, the parametric knowledge in LLMs might be limited to covering all the diverse and domain-specific queries that require grounding in various database schemas, which makes generated SQLs less accurate oftentimes. To tackle this, we propose constructing the knowledge base for text-to-SQL, a foundational source of knowledge, from which we retrieve and generate the necessary knowledge for given queries. In particular, unlike existing approaches that either manually annotate knowledge or generate only a few pieces of knowledge for each query, our knowledge base is comprehensive, which is constructed based on a combination of all the available questions and their associated database schemas along with their relevant knowledge, and can be reused for unseen databases from different datasets and domains. We validate our approach on multiple text-to-SQL datasets, considering both the overlapping and non-overlapping database scenarios, where it outperforms relevant baselines substantially.