Large model generation method and device and computer program product

Large model generation method and device and computer program product

The invention discloses a large model generation method and device and a computer program product. The method comprises the following steps: acquiring a natural language target question, and determining at least one keyword in the natural language target question; determining elements of the keyword...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: BAI HUA, BIAN JIANGTAO, FAN XIAO
Format: Patent
Sprache:chi ; eng
Schlagworte:
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
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 BAI HUA
BIAN JIANGTAO
FAN XIAO
description The invention discloses a large model generation method and device and a computer program product. The method comprises the following steps: acquiring a natural language target question, and determining at least one keyword in the natural language target question; determining elements of the keywords and attribute information of the elements by utilizing a preset knowledge graph, and generating an initial SQL statement according to the elements of the keywords and the attribute information of the elements; determining the similarity between a natural language sample question in a preset corpus and the natural language target question or between a sample SQL statement in the corpus and the initial SQL statement, and determining at least one group of target question and answer pairs based on the similarity; and training a preset initial large model by using the at least one group of target questions and answers as a prompt project to obtain a target large model. According to the method and the device, the techn
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN118427214A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN118427214A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN118427214A3</originalsourceid><addsrcrecordid>eNrjZHD1SSxKT1XIzU9JzVFIT81LLUosyczPU8hNLcnIT1FIzEtRSEkty0xOBTOT83MLSktSixQKivLTixJzQXRKaXIJDwNrWmJOcSovlOZmUHRzDXH20E0tyI9PLS5ITAaaXBLv7GdoaGFiZG5kaOJoTIwaAOh2My8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Large model generation method and device and computer program product</title><source>esp@cenet</source><creator>BAI HUA ; BIAN JIANGTAO ; FAN XIAO</creator><creatorcontrib>BAI HUA ; BIAN JIANGTAO ; FAN XIAO</creatorcontrib><description>The invention discloses a large model generation method and device and a computer program product. The method comprises the following steps: acquiring a natural language target question, and determining at least one keyword in the natural language target question; determining elements of the keywords and attribute information of the elements by utilizing a preset knowledge graph, and generating an initial SQL statement according to the elements of the keywords and the attribute information of the elements; determining the similarity between a natural language sample question in a preset corpus and the natural language target question or between a sample SQL statement in the corpus and the initial SQL statement, and determining at least one group of target question and answer pairs based on the similarity; and training a preset initial large model by using the at least one group of target questions and answers as a prompt project to obtain a target large model. According to the method and the device, the techn</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; PHYSICS</subject><creationdate>2024</creationdate><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://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240802&amp;DB=EPODOC&amp;CC=CN&amp;NR=118427214A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240802&amp;DB=EPODOC&amp;CC=CN&amp;NR=118427214A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>BAI HUA</creatorcontrib><creatorcontrib>BIAN JIANGTAO</creatorcontrib><creatorcontrib>FAN XIAO</creatorcontrib><title>Large model generation method and device and computer program product</title><description>The invention discloses a large model generation method and device and a computer program product. The method comprises the following steps: acquiring a natural language target question, and determining at least one keyword in the natural language target question; determining elements of the keywords and attribute information of the elements by utilizing a preset knowledge graph, and generating an initial SQL statement according to the elements of the keywords and the attribute information of the elements; determining the similarity between a natural language sample question in a preset corpus and the natural language target question or between a sample SQL statement in the corpus and the initial SQL statement, and determining at least one group of target question and answer pairs based on the similarity; and training a preset initial large model by using the at least one group of target questions and answers as a prompt project to obtain a target large model. According to the method and the device, the techn</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZHD1SSxKT1XIzU9JzVFIT81LLUosyczPU8hNLcnIT1FIzEtRSEkty0xOBTOT83MLSktSixQKivLTixJzQXRKaXIJDwNrWmJOcSovlOZmUHRzDXH20E0tyI9PLS5ITAaaXBLv7GdoaGFiZG5kaOJoTIwaAOh2My8</recordid><startdate>20240802</startdate><enddate>20240802</enddate><creator>BAI HUA</creator><creator>BIAN JIANGTAO</creator><creator>FAN XIAO</creator><scope>EVB</scope></search><sort><creationdate>20240802</creationdate><title>Large model generation method and device and computer program product</title><author>BAI HUA ; BIAN JIANGTAO ; FAN XIAO</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN118427214A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>BAI HUA</creatorcontrib><creatorcontrib>BIAN JIANGTAO</creatorcontrib><creatorcontrib>FAN XIAO</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>BAI HUA</au><au>BIAN JIANGTAO</au><au>FAN XIAO</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Large model generation method and device and computer program product</title><date>2024-08-02</date><risdate>2024</risdate><abstract>The invention discloses a large model generation method and device and a computer program product. The method comprises the following steps: acquiring a natural language target question, and determining at least one keyword in the natural language target question; determining elements of the keywords and attribute information of the elements by utilizing a preset knowledge graph, and generating an initial SQL statement according to the elements of the keywords and the attribute information of the elements; determining the similarity between a natural language sample question in a preset corpus and the natural language target question or between a sample SQL statement in the corpus and the initial SQL statement, and determining at least one group of target question and answer pairs based on the similarity; and training a preset initial large model by using the at least one group of target questions and answers as a prompt project to obtain a target large model. According to the method and the device, the techn</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language chi ; eng
recordid cdi_epo_espacenet_CN118427214A
source esp@cenet
subjects CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
title Large model generation method and device and computer program product
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T09%3A24%3A31IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=BAI%20HUA&rft.date=2024-08-02&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN118427214A%3C/epo_EVB%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