English

Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge

Computation and Language 2023-01-04 v1

Abstract

In this paper, we study the problem of knowledge-intensive text-to-SQL, in which domain knowledge is necessary to parse expert questions into SQL queries over domain-specific tables. We formalize this scenario by building a new Chinese benchmark KnowSQL consisting of domain-specific questions covering various domains. We then address this problem by presenting formulaic knowledge, rather than by annotating additional data examples. More concretely, we construct a formulaic knowledge bank as a domain knowledge base and propose a framework (ReGrouP) to leverage this formulaic knowledge during parsing. Experiments using ReGrouP demonstrate a significant 28.2% improvement overall on KnowSQL.

Keywords

Cite

@article{arxiv.2301.01067,
  title  = {Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge},
  author = {Longxu Dou and Yan Gao and Xuqi Liu and Mingyang Pan and Dingzirui Wang and Wanxiang Che and Dechen Zhan and Min-Yen Kan and Jian-Guang Lou},
  journal= {arXiv preprint arXiv:2301.01067},
  year   = {2023}
}

Comments

EMNLP 2022 Main Conference

R2 v1 2026-06-28T08:00:45.191Z