Multi-Matching Network for Multiple Choice Reading Comprehension

Multiple-choice machine reading comprehension is an important and challenging task where the machine is required to select the correct answer from a set of candidate answers given passage and question. Existing approaches either match extracted evidence with candidate answers shallowly or model pass...

Full description

Saved in:
Bibliographic Details
Main Authors: Tang, Min, Cai, Jiaran, Zhuo, Hankz Hankui
Format: Conference Proceeding
Language:English
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c181t-3646f0f9a2f17cc08b8a44612c2b444db80b8dfdd01c3ee11d85e7167cbed1993
cites
container_end_page 7095
container_issue 1
container_start_page 7088
container_title
container_volume 33
creator Tang, Min
Cai, Jiaran
Zhuo, Hankz Hankui
description Multiple-choice machine reading comprehension is an important and challenging task where the machine is required to select the correct answer from a set of candidate answers given passage and question. Existing approaches either match extracted evidence with candidate answers shallowly or model passage, question and candidate answers with a single paradigm of matching. In this paper, we propose Multi-Matching Network (MMN) which models the semantic relationship among passage, question and candidate answers from multiple different paradigms of matching. In our MMN model, each paradigm is inspired by how human think and designed under a unified compose-match framework. To demonstrate the effectiveness of our model, we evaluate MMN on a large-scale multiple choice machine reading comprehension dataset (i.e. RACE). Empirical results show that our proposed model achieves a significant improvement compared to strong baselines and obtains state-of-the-art results.
doi_str_mv 10.1609/aaai.v33i01.33017088
format conference_proceeding
fullrecord <record><control><sourceid>crossref</sourceid><recordid>TN_cdi_crossref_primary_10_1609_aaai_v33i01_33017088</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>10_1609_aaai_v33i01_33017088</sourcerecordid><originalsourceid>FETCH-LOGICAL-c181t-3646f0f9a2f17cc08b8a44612c2b444db80b8dfdd01c3ee11d85e7167cbed1993</originalsourceid><addsrcrecordid>eNo1kNtKxDAYhIMouKz7Bl70BVLzN2kOd0rxBLsKotchzcFGu9uSVMW3d-vq3MzADHPxIXQOpARO1IUxJpaflEYCJaUEBJHyCC0qKhimjMvjfYZa4ZoqdYpWOb-RvZgCALFAl5uPfop4Yybbxd1r8eCnryG9F2FIxW819r5ouiFaXzx54-ZNM2zH5Du_y3HYnaGTYPrsV3--RC8318_NHV4_3t43V2tsQcKEKWc8kKBMFUBYS2QrDWMcKlu1jDHXStJKF5wjYKn3AE7WXgAXtvUOlKJLxA6_Ng05Jx_0mOLWpG8NRM8g9AxCH0DofxD0B_mKUvQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Multi-Matching Network for Multiple Choice Reading Comprehension</title><source>Freely Accessible Journals</source><creator>Tang, Min ; Cai, Jiaran ; Zhuo, Hankz Hankui</creator><creatorcontrib>Tang, Min ; Cai, Jiaran ; Zhuo, Hankz Hankui</creatorcontrib><description>Multiple-choice machine reading comprehension is an important and challenging task where the machine is required to select the correct answer from a set of candidate answers given passage and question. Existing approaches either match extracted evidence with candidate answers shallowly or model passage, question and candidate answers with a single paradigm of matching. In this paper, we propose Multi-Matching Network (MMN) which models the semantic relationship among passage, question and candidate answers from multiple different paradigms of matching. In our MMN model, each paradigm is inspired by how human think and designed under a unified compose-match framework. To demonstrate the effectiveness of our model, we evaluate MMN on a large-scale multiple choice machine reading comprehension dataset (i.e. RACE). Empirical results show that our proposed model achieves a significant improvement compared to strong baselines and obtains state-of-the-art results.</description><identifier>ISSN: 2159-5399</identifier><identifier>EISSN: 2374-3468</identifier><identifier>DOI: 10.1609/aaai.v33i01.33017088</identifier><language>eng</language><ispartof>Proceedings of the ... AAAI Conference on Artificial Intelligence, 2019, Vol.33 (1), p.7088-7095</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c181t-3646f0f9a2f17cc08b8a44612c2b444db80b8dfdd01c3ee11d85e7167cbed1993</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Tang, Min</creatorcontrib><creatorcontrib>Cai, Jiaran</creatorcontrib><creatorcontrib>Zhuo, Hankz Hankui</creatorcontrib><title>Multi-Matching Network for Multiple Choice Reading Comprehension</title><title>Proceedings of the ... AAAI Conference on Artificial Intelligence</title><description>Multiple-choice machine reading comprehension is an important and challenging task where the machine is required to select the correct answer from a set of candidate answers given passage and question. Existing approaches either match extracted evidence with candidate answers shallowly or model passage, question and candidate answers with a single paradigm of matching. In this paper, we propose Multi-Matching Network (MMN) which models the semantic relationship among passage, question and candidate answers from multiple different paradigms of matching. In our MMN model, each paradigm is inspired by how human think and designed under a unified compose-match framework. To demonstrate the effectiveness of our model, we evaluate MMN on a large-scale multiple choice machine reading comprehension dataset (i.e. RACE). Empirical results show that our proposed model achieves a significant improvement compared to strong baselines and obtains state-of-the-art results.</description><issn>2159-5399</issn><issn>2374-3468</issn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2019</creationdate><recordtype>conference_proceeding</recordtype><recordid>eNo1kNtKxDAYhIMouKz7Bl70BVLzN2kOd0rxBLsKotchzcFGu9uSVMW3d-vq3MzADHPxIXQOpARO1IUxJpaflEYCJaUEBJHyCC0qKhimjMvjfYZa4ZoqdYpWOb-RvZgCALFAl5uPfop4Yybbxd1r8eCnryG9F2FIxW819r5ouiFaXzx54-ZNM2zH5Du_y3HYnaGTYPrsV3--RC8318_NHV4_3t43V2tsQcKEKWc8kKBMFUBYS2QrDWMcKlu1jDHXStJKF5wjYKn3AE7WXgAXtvUOlKJLxA6_Ng05Jx_0mOLWpG8NRM8g9AxCH0DofxD0B_mKUvQ</recordid><startdate>20190717</startdate><enddate>20190717</enddate><creator>Tang, Min</creator><creator>Cai, Jiaran</creator><creator>Zhuo, Hankz Hankui</creator><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20190717</creationdate><title>Multi-Matching Network for Multiple Choice Reading Comprehension</title><author>Tang, Min ; Cai, Jiaran ; Zhuo, Hankz Hankui</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c181t-3646f0f9a2f17cc08b8a44612c2b444db80b8dfdd01c3ee11d85e7167cbed1993</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2019</creationdate><toplevel>online_resources</toplevel><creatorcontrib>Tang, Min</creatorcontrib><creatorcontrib>Cai, Jiaran</creatorcontrib><creatorcontrib>Zhuo, Hankz Hankui</creatorcontrib><collection>CrossRef</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Tang, Min</au><au>Cai, Jiaran</au><au>Zhuo, Hankz Hankui</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Multi-Matching Network for Multiple Choice Reading Comprehension</atitle><btitle>Proceedings of the ... AAAI Conference on Artificial Intelligence</btitle><date>2019-07-17</date><risdate>2019</risdate><volume>33</volume><issue>1</issue><spage>7088</spage><epage>7095</epage><pages>7088-7095</pages><issn>2159-5399</issn><eissn>2374-3468</eissn><abstract>Multiple-choice machine reading comprehension is an important and challenging task where the machine is required to select the correct answer from a set of candidate answers given passage and question. Existing approaches either match extracted evidence with candidate answers shallowly or model passage, question and candidate answers with a single paradigm of matching. In this paper, we propose Multi-Matching Network (MMN) which models the semantic relationship among passage, question and candidate answers from multiple different paradigms of matching. In our MMN model, each paradigm is inspired by how human think and designed under a unified compose-match framework. To demonstrate the effectiveness of our model, we evaluate MMN on a large-scale multiple choice machine reading comprehension dataset (i.e. RACE). Empirical results show that our proposed model achieves a significant improvement compared to strong baselines and obtains state-of-the-art results.</abstract><doi>10.1609/aaai.v33i01.33017088</doi><tpages>8</tpages></addata></record>
fulltext fulltext
identifier ISSN: 2159-5399
ispartof Proceedings of the ... AAAI Conference on Artificial Intelligence, 2019, Vol.33 (1), p.7088-7095
issn 2159-5399
2374-3468
language eng
recordid cdi_crossref_primary_10_1609_aaai_v33i01_33017088
source Freely Accessible Journals
title Multi-Matching Network for Multiple Choice Reading Comprehension
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-06T20%3A43%3A48IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-crossref&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Multi-Matching%20Network%20for%20Multiple%20Choice%20Reading%20Comprehension&rft.btitle=Proceedings%20of%20the%20...%20AAAI%20Conference%20on%20Artificial%20Intelligence&rft.au=Tang,%20Min&rft.date=2019-07-17&rft.volume=33&rft.issue=1&rft.spage=7088&rft.epage=7095&rft.pages=7088-7095&rft.issn=2159-5399&rft.eissn=2374-3468&rft_id=info:doi/10.1609/aaai.v33i01.33017088&rft_dat=%3Ccrossref%3E10_1609_aaai_v33i01_33017088%3C/crossref%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c181t-3646f0f9a2f17cc08b8a44612c2b444db80b8dfdd01c3ee11d85e7167cbed1993%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true