You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL

While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YOR...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-09
Main Authors: Kobayashi, Hideo, Lan, Wuwei, Shi, Peng, Chang, Shuaichen, Guo, Jiang, Zhu, Henghui, Wang, Zhiguo, Ng, Patrick
Format: Article
Language:English
Subjects:
Inference
Query languages
Questions
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Kobayashi, Hideo
Lan, Wuwei
Shi, Peng
Chang, Shuaichen
Guo, Jiang
Zhu, Henghui
Wang, Zhiguo
Ng, Patrick
description While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_3106850019</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3106850019</sourcerecordid><originalsourceid>FETCH-proquest_journals_31068500193</originalsourceid><addsrcrecordid>eNqNjMsKgkAUQIcgSMp_uNCmFsI4k2Zte1AUSGYLVzLlVRSZKWekx9dX0Ae0OmdxOB1iMc5dJ5gw1iO21hWllPlT5nncIqdEtRDK-gkRiuxjF4RREkbheA57FI0sZQFGwVYabKSoyxfCUhhxFhphJ9W9xqxAyFUDMT6MY5RzPOwHpJuLWqP9Y58M16t4sXGujbq1qE1aqfa70yl3qR94lLoz_l_1BokuPpY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3106850019</pqid></control><display><type>article</type><title>You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL</title><source>Publicly Available Content Database</source><creator>Kobayashi, Hideo ; Lan, Wuwei ; Shi, Peng ; Chang, Shuaichen ; Guo, Jiang ; Zhu, Henghui ; Wang, Zhiguo ; Ng, Patrick</creator><creatorcontrib>Kobayashi, Hideo ; Lan, Wuwei ; Shi, Peng ; Chang, Shuaichen ; Guo, Jiang ; Zhu, Henghui ; Wang, Zhiguo ; Ng, Patrick</creatorcontrib><description>While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Inference ; Query languages ; Questions</subject><ispartof>arXiv.org, 2024-09</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by-sa/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/3106850019?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Kobayashi, Hideo</creatorcontrib><creatorcontrib>Lan, Wuwei</creatorcontrib><creatorcontrib>Shi, Peng</creatorcontrib><creatorcontrib>Chang, Shuaichen</creatorcontrib><creatorcontrib>Guo, Jiang</creatorcontrib><creatorcontrib>Zhu, Henghui</creatorcontrib><creatorcontrib>Wang, Zhiguo</creatorcontrib><creatorcontrib>Ng, Patrick</creatorcontrib><title>You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL</title><title>arXiv.org</title><description>While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation.</description><subject>Inference</subject><subject>Query languages</subject><subject>Questions</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNjMsKgkAUQIcgSMp_uNCmFsI4k2Zte1AUSGYLVzLlVRSZKWekx9dX0Ae0OmdxOB1iMc5dJ5gw1iO21hWllPlT5nncIqdEtRDK-gkRiuxjF4RREkbheA57FI0sZQFGwVYabKSoyxfCUhhxFhphJ9W9xqxAyFUDMT6MY5RzPOwHpJuLWqP9Y58M16t4sXGujbq1qE1aqfa70yl3qR94lLoz_l_1BokuPpY</recordid><startdate>20240918</startdate><enddate>20240918</enddate><creator>Kobayashi, Hideo</creator><creator>Lan, Wuwei</creator><creator>Shi, Peng</creator><creator>Chang, Shuaichen</creator><creator>Guo, Jiang</creator><creator>Zhu, Henghui</creator><creator>Wang, Zhiguo</creator><creator>Ng, Patrick</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20240918</creationdate><title>You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL</title><author>Kobayashi, Hideo ; Lan, Wuwei ; Shi, Peng ; Chang, Shuaichen ; Guo, Jiang ; Zhu, Henghui ; Wang, Zhiguo ; Ng, Patrick</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_31068500193</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Inference</topic><topic>Query languages</topic><topic>Questions</topic><toplevel>online_resources</toplevel><creatorcontrib>Kobayashi, Hideo</creatorcontrib><creatorcontrib>Lan, Wuwei</creatorcontrib><creatorcontrib>Shi, Peng</creatorcontrib><creatorcontrib>Chang, Shuaichen</creatorcontrib><creatorcontrib>Guo, Jiang</creatorcontrib><creatorcontrib>Zhu, Henghui</creatorcontrib><creatorcontrib>Wang, Zhiguo</creatorcontrib><creatorcontrib>Ng, Patrick</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kobayashi, Hideo</au><au>Lan, Wuwei</au><au>Shi, Peng</au><au>Chang, Shuaichen</au><au>Guo, Jiang</au><au>Zhu, Henghui</au><au>Wang, Zhiguo</au><au>Ng, Patrick</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL</atitle><jtitle>arXiv.org</jtitle><date>2024-09-18</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2024-09
issn 2331-8422
language eng
recordid cdi_proquest_journals_3106850019
source Publicly Available Content Database
subjects Inference
Query languages
Questions
title You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T17%3A50%3A19IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=You%20Only%20Read%20Once%20(YORO):%20Learning%20to%20Internalize%20Database%20Knowledge%20for%20Text-to-SQL&rft.jtitle=arXiv.org&rft.au=Kobayashi,%20Hideo&rft.date=2024-09-18&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E3106850019%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_31068500193%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=3106850019&rft_id=info:pmid/&rfr_iscdi=true