Syntax-aware Natural Language Inference with Graph Matching Networks

The task of entailment judgment aims to determine whether a hypothesis is true (entailment), false (contradiction), or undetermined (neutral) given a premise. While previous methods strike successful in several benchmarks and even exceed the human baseline, recent researches show that it remains arg...

Full description

Saved in:
Bibliographic Details
Main Authors: Lin, Yan-Tong, Wu, Meng-Tse, Su, Keh-Yih
Format: Conference Proceeding
Language:English
Subjects:
Artificial intelligence
Benchmark testing
Context modeling
dependency tree
Encoding
graph neural networks
natural language inference
Natural languages
recognize textual entailment
Task analysis
Visualization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 90
container_issue
container_start_page 85
container_title
container_volume
creator Lin, Yan-Tong
Wu, Meng-Tse
Su, Keh-Yih
description The task of entailment judgment aims to determine whether a hypothesis is true (entailment), false (contradiction), or undetermined (neutral) given a premise. While previous methods strike successful in several benchmarks and even exceed the human baseline, recent researches show that it remains arguable if the methods learn the statistical bias in the datasets. In this paper, we propose the syntax-aware Natural Language Inference (SynNLI) model, which utilizes graph matching networks to obtain syntax-guided contextualized representation while aligning the premise and the hypothesis accordingly. We show that the proposed method outperforms multiple baseline models on MNLI develop set, and visualize the model internal behavior.
doi_str_mv 10.1109/TAAI51410.2020.00024
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9382461</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9382461</ieee_id><sourcerecordid>9382461</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-af6a3b06465cde5d52e4b0aacc44fae5bf079ca65bf4e5dce442d97273da7d433</originalsourceid><addsrcrecordid>eNotj81OwzAQhA0SElXJE8DBL5CytjdOcowKlEihHCjnauNsfqCEykkV-vZEgtPM6BuNNELcKVgpBen9LsvySOEcNWhYAYDGCxGkcaKsjRBMAupSLLSJbWgTjdciGIaPuWY0oErUQjy8nfuRfkKayLPc0njydJAF9c2JGpZ5X7Pn3rGcurGVG0_HVr7Q6Nqub-SWx-nbfw434qqmw8DBvy7F-9Pjbv0cFq-bfJ0VYafBjCHVlkwJFm3kKo6qSDOWQOQcYk0clTXEqSM7G5yxY0RdpbGOTUVxhcYsxe3fbsfM-6Pvvsif96mZn1llfgGjSE0k</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Syntax-aware Natural Language Inference with Graph Matching Networks</title><source>IEEE Xplore All Conference Series</source><creator>Lin, Yan-Tong ; Wu, Meng-Tse ; Su, Keh-Yih</creator><creatorcontrib>Lin, Yan-Tong ; Wu, Meng-Tse ; Su, Keh-Yih</creatorcontrib><description>The task of entailment judgment aims to determine whether a hypothesis is true (entailment), false (contradiction), or undetermined (neutral) given a premise. While previous methods strike successful in several benchmarks and even exceed the human baseline, recent researches show that it remains arguable if the methods learn the statistical bias in the datasets. In this paper, we propose the syntax-aware Natural Language Inference (SynNLI) model, which utilizes graph matching networks to obtain syntax-guided contextualized representation while aligning the premise and the hypothesis accordingly. We show that the proposed method outperforms multiple baseline models on MNLI develop set, and visualize the model internal behavior.</description><identifier>EISSN: 2376-6824</identifier><identifier>EISBN: 9781665403801</identifier><identifier>EISBN: 1665403802</identifier><identifier>DOI: 10.1109/TAAI51410.2020.00024</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Artificial intelligence ; Benchmark testing ; Context modeling ; dependency tree ; Encoding ; graph neural networks ; natural language inference ; Natural languages ; recognize textual entailment ; Task analysis ; Visualization</subject><ispartof>2020 International Conference on Technologies and Applications of Artificial Intelligence (TAAI), 2020, p.85-90</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9382461$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,777,781,786,787,27906,54536,54913</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9382461$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Lin, Yan-Tong</creatorcontrib><creatorcontrib>Wu, Meng-Tse</creatorcontrib><creatorcontrib>Su, Keh-Yih</creatorcontrib><title>Syntax-aware Natural Language Inference with Graph Matching Networks</title><title>2020 International Conference on Technologies and Applications of Artificial Intelligence (TAAI)</title><addtitle>TAAI</addtitle><description>The task of entailment judgment aims to determine whether a hypothesis is true (entailment), false (contradiction), or undetermined (neutral) given a premise. While previous methods strike successful in several benchmarks and even exceed the human baseline, recent researches show that it remains arguable if the methods learn the statistical bias in the datasets. In this paper, we propose the syntax-aware Natural Language Inference (SynNLI) model, which utilizes graph matching networks to obtain syntax-guided contextualized representation while aligning the premise and the hypothesis accordingly. We show that the proposed method outperforms multiple baseline models on MNLI develop set, and visualize the model internal behavior.</description><subject>Artificial intelligence</subject><subject>Benchmark testing</subject><subject>Context modeling</subject><subject>dependency tree</subject><subject>Encoding</subject><subject>graph neural networks</subject><subject>natural language inference</subject><subject>Natural languages</subject><subject>recognize textual entailment</subject><subject>Task analysis</subject><subject>Visualization</subject><issn>2376-6824</issn><isbn>9781665403801</isbn><isbn>1665403802</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2020</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotj81OwzAQhA0SElXJE8DBL5CytjdOcowKlEihHCjnauNsfqCEykkV-vZEgtPM6BuNNELcKVgpBen9LsvySOEcNWhYAYDGCxGkcaKsjRBMAupSLLSJbWgTjdciGIaPuWY0oErUQjy8nfuRfkKayLPc0njydJAF9c2JGpZ5X7Pn3rGcurGVG0_HVr7Q6Nqub-SWx-nbfw434qqmw8DBvy7F-9Pjbv0cFq-bfJ0VYafBjCHVlkwJFm3kKo6qSDOWQOQcYk0clTXEqSM7G5yxY0RdpbGOTUVxhcYsxe3fbsfM-6Pvvsif96mZn1llfgGjSE0k</recordid><startdate>202012</startdate><enddate>202012</enddate><creator>Lin, Yan-Tong</creator><creator>Wu, Meng-Tse</creator><creator>Su, Keh-Yih</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>202012</creationdate><title>Syntax-aware Natural Language Inference with Graph Matching Networks</title><author>Lin, Yan-Tong ; Wu, Meng-Tse ; Su, Keh-Yih</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-af6a3b06465cde5d52e4b0aacc44fae5bf079ca65bf4e5dce442d97273da7d433</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Artificial intelligence</topic><topic>Benchmark testing</topic><topic>Context modeling</topic><topic>dependency tree</topic><topic>Encoding</topic><topic>graph neural networks</topic><topic>natural language inference</topic><topic>Natural languages</topic><topic>recognize textual entailment</topic><topic>Task analysis</topic><topic>Visualization</topic><toplevel>online_resources</toplevel><creatorcontrib>Lin, Yan-Tong</creatorcontrib><creatorcontrib>Wu, Meng-Tse</creatorcontrib><creatorcontrib>Su, Keh-Yih</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Lin, Yan-Tong</au><au>Wu, Meng-Tse</au><au>Su, Keh-Yih</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Syntax-aware Natural Language Inference with Graph Matching Networks</atitle><btitle>2020 International Conference on Technologies and Applications of Artificial Intelligence (TAAI)</btitle><stitle>TAAI</stitle><date>2020-12</date><risdate>2020</risdate><spage>85</spage><epage>90</epage><pages>85-90</pages><eissn>2376-6824</eissn><eisbn>9781665403801</eisbn><eisbn>1665403802</eisbn><coden>IEEPAD</coden><abstract>The task of entailment judgment aims to determine whether a hypothesis is true (entailment), false (contradiction), or undetermined (neutral) given a premise. While previous methods strike successful in several benchmarks and even exceed the human baseline, recent researches show that it remains arguable if the methods learn the statistical bias in the datasets. In this paper, we propose the syntax-aware Natural Language Inference (SynNLI) model, which utilizes graph matching networks to obtain syntax-guided contextualized representation while aligning the premise and the hypothesis accordingly. We show that the proposed method outperforms multiple baseline models on MNLI develop set, and visualize the model internal behavior.</abstract><pub>IEEE</pub><doi>10.1109/TAAI51410.2020.00024</doi><tpages>6</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2376-6824
ispartof 2020 International Conference on Technologies and Applications of Artificial Intelligence (TAAI), 2020, p.85-90
issn 2376-6824
language eng
recordid cdi_ieee_primary_9382461
source IEEE Xplore All Conference Series
subjects Artificial intelligence
Benchmark testing
Context modeling
dependency tree
Encoding
graph neural networks
natural language inference
Natural languages
recognize textual entailment
Task analysis
Visualization
title Syntax-aware Natural Language Inference with Graph Matching Networks
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T22%3A37%3A23IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Syntax-aware%20Natural%20Language%20Inference%20with%20Graph%20Matching%20Networks&rft.btitle=2020%20International%20Conference%20on%20Technologies%20and%20Applications%20of%20Artificial%20Intelligence%20(TAAI)&rft.au=Lin,%20Yan-Tong&rft.date=2020-12&rft.spage=85&rft.epage=90&rft.pages=85-90&rft.eissn=2376-6824&rft.coden=IEEPAD&rft_id=info:doi/10.1109/TAAI51410.2020.00024&rft.eisbn=9781665403801&rft.eisbn_list=1665403802&rft_dat=%3Cieee_CHZPO%3E9382461%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i203t-af6a3b06465cde5d52e4b0aacc44fae5bf079ca65bf4e5dce442d97273da7d433%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=9382461&rfr_iscdi=true