CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions

Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Hanchong, Cao, Ruisheng, Xu, Hongshen, Chen, Lu, Yu, Kai
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Zhang, Hanchong
Cao, Ruisheng
Xu, Hongshen
Chen, Lu
Yu, Kai
description Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In the conversational context, the current SQL query can be modified from the preceding SQL query with only a few operations due to the context dependency. We introduce our method called CoE-SQL which can prompt LLMs to generate the SQL query based on the previously generated SQL query with an edition chain. We also conduct extensive ablation studies to determine the optimal configuration of our approach. Our approach outperforms different in-context learning baselines stably and achieves state-of-the-art performances on two benchmarks SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned models.
doi_str_mv 10.48550/arxiv.2405.02712
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2405_02712</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2405_02712</sourcerecordid><originalsourceid>FETCH-LOGICAL-a672-cd20a1708cedd32b0e93c529244787a7d3a95c2a177fbc4a76d6e838a7a37aa53</originalsourceid><addsrcrecordid>eNotj71OwzAURr0woMIDMOEXcHD8k-uwoShApQBCZI9ubYdaKjZyXShvT1uYvuE7OtIh5KrmlTJa8xvM-_BVCcV1xQXU4pw8d6lnb6_DLV1G1qVY_L7QwWOOIb7TOWX6tNuUwMZdjnQ8nKykI0-_Q1nTbo0hsjSz3oUSUtxekLMZN1t_-b8LMt73Y_fIhpeHZXc3MGxAMOsExxq4sd45KVbct9Jq0QqlwACCk9hqKw4IzCurEBrXeCMNAkpA1HJBrv-0p6DpM4cPzD_TMWw6hclfQYBHHg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions</title><source>arXiv.org</source><creator>Zhang, Hanchong ; Cao, Ruisheng ; Xu, Hongshen ; Chen, Lu ; Yu, Kai</creator><creatorcontrib>Zhang, Hanchong ; Cao, Ruisheng ; Xu, Hongshen ; Chen, Lu ; Yu, Kai</creatorcontrib><description>Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In the conversational context, the current SQL query can be modified from the preceding SQL query with only a few operations due to the context dependency. We introduce our method called CoE-SQL which can prompt LLMs to generate the SQL query based on the previously generated SQL query with an edition chain. We also conduct extensive ablation studies to determine the optimal configuration of our approach. Our approach outperforms different in-context learning baselines stably and achieves state-of-the-art performances on two benchmarks SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned models.</description><identifier>DOI: 10.48550/arxiv.2405.02712</identifier><language>eng</language><subject>Computer Science - Computation and Language</subject><creationdate>2024-05</creationdate><rights>http://creativecommons.org/licenses/by-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2405.02712$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2405.02712$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhang, Hanchong</creatorcontrib><creatorcontrib>Cao, Ruisheng</creatorcontrib><creatorcontrib>Xu, Hongshen</creatorcontrib><creatorcontrib>Chen, Lu</creatorcontrib><creatorcontrib>Yu, Kai</creatorcontrib><title>CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions</title><description>Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In the conversational context, the current SQL query can be modified from the preceding SQL query with only a few operations due to the context dependency. We introduce our method called CoE-SQL which can prompt LLMs to generate the SQL query based on the previously generated SQL query with an edition chain. We also conduct extensive ablation studies to determine the optimal configuration of our approach. Our approach outperforms different in-context learning baselines stably and achieves state-of-the-art performances on two benchmarks SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned models.</description><subject>Computer Science - Computation and Language</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj71OwzAURr0woMIDMOEXcHD8k-uwoShApQBCZI9ubYdaKjZyXShvT1uYvuE7OtIh5KrmlTJa8xvM-_BVCcV1xQXU4pw8d6lnb6_DLV1G1qVY_L7QwWOOIb7TOWX6tNuUwMZdjnQ8nKykI0-_Q1nTbo0hsjSz3oUSUtxekLMZN1t_-b8LMt73Y_fIhpeHZXc3MGxAMOsExxq4sd45KVbct9Jq0QqlwACCk9hqKw4IzCurEBrXeCMNAkpA1HJBrv-0p6DpM4cPzD_TMWw6hclfQYBHHg</recordid><startdate>20240504</startdate><enddate>20240504</enddate><creator>Zhang, Hanchong</creator><creator>Cao, Ruisheng</creator><creator>Xu, Hongshen</creator><creator>Chen, Lu</creator><creator>Yu, Kai</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240504</creationdate><title>CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions</title><author>Zhang, Hanchong ; Cao, Ruisheng ; Xu, Hongshen ; Chen, Lu ; Yu, Kai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a672-cd20a1708cedd32b0e93c529244787a7d3a95c2a177fbc4a76d6e838a7a37aa53</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computation and Language</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Hanchong</creatorcontrib><creatorcontrib>Cao, Ruisheng</creatorcontrib><creatorcontrib>Xu, Hongshen</creatorcontrib><creatorcontrib>Chen, Lu</creatorcontrib><creatorcontrib>Yu, Kai</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhang, Hanchong</au><au>Cao, Ruisheng</au><au>Xu, Hongshen</au><au>Chen, Lu</au><au>Yu, Kai</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions</atitle><date>2024-05-04</date><risdate>2024</risdate><abstract>Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In the conversational context, the current SQL query can be modified from the preceding SQL query with only a few operations due to the context dependency. We introduce our method called CoE-SQL which can prompt LLMs to generate the SQL query based on the previously generated SQL query with an edition chain. We also conduct extensive ablation studies to determine the optimal configuration of our approach. Our approach outperforms different in-context learning baselines stably and achieves state-of-the-art performances on two benchmarks SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned models.</abstract><doi>10.48550/arxiv.2405.02712</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2405.02712
ispartof
issn
language eng
recordid cdi_arxiv_primary_2405_02712
source arXiv.org
subjects Computer Science - Computation and Language
title CoE-SQL: In-Context Learning for Multi-Turn Text-to-SQL with Chain-of-Editions
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T23%3A48%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=CoE-SQL:%20In-Context%20Learning%20for%20Multi-Turn%20Text-to-SQL%20with%20Chain-of-Editions&rft.au=Zhang,%20Hanchong&rft.date=2024-05-04&rft_id=info:doi/10.48550/arxiv.2405.02712&rft_dat=%3Carxiv_GOX%3E2405_02712%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true