@inproceedings{zhang-etal-2023-refsql,
title = "{R}e{FSQL}: A Retrieval-Augmentation Framework for Text-to-{SQL} Generation",
author = "Zhang, Kun and
Lin, Xiexiong and
Wang, Yuanzhuo and
Zhang, Xin and
Sun, Fei and
Jianhe, Cen and
Tan, Hexiang and
Jiang, Xuhui and
Shen, Huawei",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2023",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-emnlp.48/",
doi = "10.18653/v1/2023.findings-emnlp.48",
pages = "664--673",
abstract = "Text-to-SQL is the task that aims at translating natural language questions into SQL queries. Existing methods directly align the natural language with SQL Language and train one encoder-decoder-based model to fit all questions. However, they underestimate the inherent structural characteristics of SQL, as well as the gap between specific structure knowledge and general knowledge. This leads to structure errors in the generated SQL. To address the above challenges, we propose a retrieval-argument framework, namely ReFSQL. It contains two parts, structure-enhanced retriever and the generator. Structure-enhanced retriever is designed to identify samples with comparable specific knowledge in an unsupervised way. Subsequently, we incorporate the retrieved samples' SQL into the input, enabling the model to acquire prior knowledge of similar SQL grammar. To further bridge the gap between specific and general knowledge, we present a mahalanobis contrastive learning method, which facilitates the transfer of the sample toward the specific knowledge distribution constructed by the retrieved samples. Experimental results on five datasets verify the effectiveness of our approach in improving the accuracy and robustness of Text-to-SQL generation. Our framework has achieved improved performance when combined with many other backbone models (including the 11B flan-T5) and also achieved state-of-the-art performance when compared to existing methods that employ the fine-tuning approach."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2023-refsql">
<titleInfo>
<title>ReFSQL: A Retrieval-Augmentation Framework for Text-to-SQL Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiexiong</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanzhuo</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cen</namePart>
<namePart type="family">Jianhe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hexiang</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuhui</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huawei</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="family">Pino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Text-to-SQL is the task that aims at translating natural language questions into SQL queries. Existing methods directly align the natural language with SQL Language and train one encoder-decoder-based model to fit all questions. However, they underestimate the inherent structural characteristics of SQL, as well as the gap between specific structure knowledge and general knowledge. This leads to structure errors in the generated SQL. To address the above challenges, we propose a retrieval-argument framework, namely ReFSQL. It contains two parts, structure-enhanced retriever and the generator. Structure-enhanced retriever is designed to identify samples with comparable specific knowledge in an unsupervised way. Subsequently, we incorporate the retrieved samples’ SQL into the input, enabling the model to acquire prior knowledge of similar SQL grammar. To further bridge the gap between specific and general knowledge, we present a mahalanobis contrastive learning method, which facilitates the transfer of the sample toward the specific knowledge distribution constructed by the retrieved samples. Experimental results on five datasets verify the effectiveness of our approach in improving the accuracy and robustness of Text-to-SQL generation. Our framework has achieved improved performance when combined with many other backbone models (including the 11B flan-T5) and also achieved state-of-the-art performance when compared to existing methods that employ the fine-tuning approach.</abstract>
<identifier type="citekey">zhang-etal-2023-refsql</identifier>
<identifier type="doi">10.18653/v1/2023.findings-emnlp.48</identifier>
<location>
<url>https://aclanthology.org/2023.findings-emnlp.48/</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>664</start>
<end>673</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ReFSQL: A Retrieval-Augmentation Framework for Text-to-SQL Generation
%A Zhang, Kun
%A Lin, Xiexiong
%A Wang, Yuanzhuo
%A Zhang, Xin
%A Sun, Fei
%A Jianhe, Cen
%A Tan, Hexiang
%A Jiang, Xuhui
%A Shen, Huawei
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Findings of the Association for Computational Linguistics: EMNLP 2023
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F zhang-etal-2023-refsql
%X Text-to-SQL is the task that aims at translating natural language questions into SQL queries. Existing methods directly align the natural language with SQL Language and train one encoder-decoder-based model to fit all questions. However, they underestimate the inherent structural characteristics of SQL, as well as the gap between specific structure knowledge and general knowledge. This leads to structure errors in the generated SQL. To address the above challenges, we propose a retrieval-argument framework, namely ReFSQL. It contains two parts, structure-enhanced retriever and the generator. Structure-enhanced retriever is designed to identify samples with comparable specific knowledge in an unsupervised way. Subsequently, we incorporate the retrieved samples’ SQL into the input, enabling the model to acquire prior knowledge of similar SQL grammar. To further bridge the gap between specific and general knowledge, we present a mahalanobis contrastive learning method, which facilitates the transfer of the sample toward the specific knowledge distribution constructed by the retrieved samples. Experimental results on five datasets verify the effectiveness of our approach in improving the accuracy and robustness of Text-to-SQL generation. Our framework has achieved improved performance when combined with many other backbone models (including the 11B flan-T5) and also achieved state-of-the-art performance when compared to existing methods that employ the fine-tuning approach.
%R 10.18653/v1/2023.findings-emnlp.48
%U https://aclanthology.org/2023.findings-emnlp.48/
%U https://doi.org/10.18653/v1/2023.findings-emnlp.48
%P 664-673
Markdown (Informal)
[ReFSQL: A Retrieval-Augmentation Framework for Text-to-SQL Generation](https://aclanthology.org/2023.findings-emnlp.48/) (Zhang et al., Findings 2023)
ACL
- Kun Zhang, Xiexiong Lin, Yuanzhuo Wang, Xin Zhang, Fei Sun, Cen Jianhe, Hexiang Tan, Xuhui Jiang, and Huawei Shen. 2023. ReFSQL: A Retrieval-Augmentation Framework for Text-to-SQL Generation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 664–673, Singapore. Association for Computational Linguistics.