Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning

A common problem with adopting Text-to-SQL translation in database systems is poor generalization. Specifically, when there is limited training data on new datasets, existing few-shot Text-to-SQL techniques, even with carefully designed textual prompts on pre-trained language models (PLMs), tend to...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the ACM on management of data 2023-06, Vol.1 (2), p.1-28, Article 147
Main Authors: Gu, Zihui, Fan, Ju, Tang, Nan, Cao, Lei, Jia, Bowen, Madden, Sam, Du, Xiaoyong
Format: Article
Language:English
Subjects:
Data management systems
Information systems
Parsing
Program reasoning
Query languages
Relational database query languages
Semantics and reasoning
Theory of computation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A common problem with adopting Text-to-SQL translation in database systems is poor generalization. Specifically, when there is limited training data on new datasets, existing few-shot Text-to-SQL techniques, even with carefully designed textual prompts on pre-trained language models (PLMs), tend to be ineffective. In this paper, we present a divide-and-conquer framework to better support few-shot Text-to-SQL translation, which divides Text-to-SQL translation into two stages (or sub-tasks), such that each sub-task is simpler to be tackled. The first stage, called the structure stage, steers a PLM to generate an SQL structure (including SQL commands such as SELECT, FROM, WHERE and SQL operators such as
ISSN:2836-6573
2836-6573
DOI:10.1145/3589292