A Survey on Employing Large Language Models for Text-to-SQL Tasks

As the volume of data continues to increase, the capability to efficiently query and leverage this data has emerged as a pivotal factor in enhancing competitiveness across numerous sectors in this era. Relational databases require the use of SQL for querying. However, writing SQL necessitates specialized knowledge, which creates barriers for unprofessional users to query and access databases. Text-to-SQL parsing is a well-established task in the field of natural language processing (NLP). Its purpose is to convert natural language queries into SQL queries, bridging the gap between non-expert users and database access. To illustrate, imagine a table named cities with three columns: city_name (type: string), population (type: integer), and country (type: string). If we are given the natural language query "Find all the cities with a population greater than 1 million in the United States," the Text-to-SQL parsing technique should automatically generate the correct SQL query: Both authors contributed equally to this research. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page.