TypeSQL: Knowledge-based Type-Aware Neural Text-to-SQL Generation

Tao Yu, Zifan Li, Zilin Zhang, Rui Zhang, Dragomir Radev

2018-04-25NAACL 2018 6Text-To-SQL slot-filling Slot Filling Vocal Bursts Type Prediction

Abstract

Interacting with relational databases through natural language helps users of any background easily query and analyze a vast amount of data. This requires a system that understands users' questions and converts them to SQL queries automatically. In this paper we present a novel approach, TypeSQL, which views this problem as a slot filling task. Additionally, TypeSQL utilizes type information to better understand rare entities and numbers in natural language questions. We test this idea on the WikiSQL dataset and outperform the prior state-of-the-art by 5.5% in much less time. We also show that accessing the content of databases can significantly improve the performance when users' queries are not well-formed. TypeSQL gets 82.6% accuracy, a 17.5% absolute improvement compared to the previous content-sensitive model.

Results

Task	Dataset	Metric	Value	Model
Code Generation	WikiSQL	Execution Accuracy	82.6	TypeSQL+TC (Yu et al., 2018)+
Code Generation	WikiSQL	Execution Accuracy	73.5	TypeSQL (Yu et al., 2018)

Related Papers

CogniSQL-R1-Zero: Lightweight Reinforced Reasoning for Efficient SQL Generation2025-07-08 XiYan-SQL: A Novel Multi-Generator Framework For Text-to-SQL2025-07-07 SWE-SQL: Illuminating LLM Pathways to Solve User SQL Issues in Real-World Applications2025-06-23 Schema-R1: A reasoning training approach for schema linking in Text-to-SQL Task2025-06-13 Invocable APIs derived from NL2SQL datasets for LLM Tool-Calling Evaluation2025-06-12 Bridging the Gap Between Open-Source and Proprietary LLMs in Table QA2025-06-11 LLM-Driven Data Generation and a Novel Soft Metric for Evaluating Text-to-SQL in Aviation MRO2025-06-11 HI-SQL: Optimizing Text-to-SQL Systems through Dynamic Hint Integration2025-06-11