Template Filling for Controllable Commonsense Reasoning
Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2022-10 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Rajagopal, Dheeraj Khetan, Vivek Sacaleanu, Bogdan Gershman, Anatole Fano, Andrew Hovy, Eduard |
description | Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study the commonsense reasoning as a template filling task (TemplateCSR) -- where the language models fills reasoning templates with the given constraints as control factors. As an approach to TemplateCSR, we (i) propose a dataset of commonsense reasoning template-expansion pairs and (ii) introduce POTTER, a pretrained sequence-to-sequence model using prompts to perform commonsense reasoning across concepts. Our experiments show that our approach outperforms baselines both in generation metrics and factuality metrics. We also present a detailed error analysis on our approach's ability to reliably perform commonsense reasoning. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2591842709</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2591842709</sourcerecordid><originalsourceid>FETCH-proquest_journals_25918427093</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwD0nNLchJLElVcMvMycnMS1dIyy9ScM7PKynKz8lJTMpJBXJyc_PzilOBSCEoNbE4Pw-ojIeBNS0xpziVF0pzMyi7uYY4e-gWFOUXlqYWl8Rn5ZcW5QGl4o1MLQ2BlpsbWBoTpwoAKjc2EA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2591842709</pqid></control><display><type>article</type><title>Template Filling for Controllable Commonsense Reasoning</title><source>Freely Accessible Journals</source><creator>Rajagopal, Dheeraj ; Khetan, Vivek ; Sacaleanu, Bogdan ; Gershman, Anatole ; Fano, Andrew ; Hovy, Eduard</creator><creatorcontrib>Rajagopal, Dheeraj ; Khetan, Vivek ; Sacaleanu, Bogdan ; Gershman, Anatole ; Fano, Andrew ; Hovy, Eduard</creatorcontrib><description>Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study the commonsense reasoning as a template filling task (TemplateCSR) -- where the language models fills reasoning templates with the given constraints as control factors. As an approach to TemplateCSR, we (i) propose a dataset of commonsense reasoning template-expansion pairs and (ii) introduce POTTER, a pretrained sequence-to-sequence model using prompts to perform commonsense reasoning across concepts. Our experiments show that our approach outperforms baselines both in generation metrics and factuality metrics. We also present a detailed error analysis on our approach's ability to reliably perform commonsense reasoning.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Coders ; Domains ; Encoders-Decoders ; Error analysis ; Reasoning</subject><ispartof>arXiv.org, 2022-10</ispartof><rights>2022. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (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><link.rule.ids>780,784</link.rule.ids></links><search><creatorcontrib>Rajagopal, Dheeraj</creatorcontrib><creatorcontrib>Khetan, Vivek</creatorcontrib><creatorcontrib>Sacaleanu, Bogdan</creatorcontrib><creatorcontrib>Gershman, Anatole</creatorcontrib><creatorcontrib>Fano, Andrew</creatorcontrib><creatorcontrib>Hovy, Eduard</creatorcontrib><title>Template Filling for Controllable Commonsense Reasoning</title><title>arXiv.org</title><description>Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study the commonsense reasoning as a template filling task (TemplateCSR) -- where the language models fills reasoning templates with the given constraints as control factors. As an approach to TemplateCSR, we (i) propose a dataset of commonsense reasoning template-expansion pairs and (ii) introduce POTTER, a pretrained sequence-to-sequence model using prompts to perform commonsense reasoning across concepts. Our experiments show that our approach outperforms baselines both in generation metrics and factuality metrics. We also present a detailed error analysis on our approach's ability to reliably perform commonsense reasoning.</description><subject>Coders</subject><subject>Domains</subject><subject>Encoders-Decoders</subject><subject>Error analysis</subject><subject>Reasoning</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwD0nNLchJLElVcMvMycnMS1dIyy9ScM7PKynKz8lJTMpJBXJyc_PzilOBSCEoNbE4Pw-ojIeBNS0xpziVF0pzMyi7uYY4e-gWFOUXlqYWl8Rn5ZcW5QGl4o1MLQ2BlpsbWBoTpwoAKjc2EA</recordid><startdate>20221014</startdate><enddate>20221014</enddate><creator>Rajagopal, Dheeraj</creator><creator>Khetan, Vivek</creator><creator>Sacaleanu, Bogdan</creator><creator>Gershman, Anatole</creator><creator>Fano, Andrew</creator><creator>Hovy, Eduard</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>20221014</creationdate><title>Template Filling for Controllable Commonsense Reasoning</title><author>Rajagopal, Dheeraj ; Khetan, Vivek ; Sacaleanu, Bogdan ; Gershman, Anatole ; Fano, Andrew ; Hovy, Eduard</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_25918427093</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Coders</topic><topic>Domains</topic><topic>Encoders-Decoders</topic><topic>Error analysis</topic><topic>Reasoning</topic><toplevel>online_resources</toplevel><creatorcontrib>Rajagopal, Dheeraj</creatorcontrib><creatorcontrib>Khetan, Vivek</creatorcontrib><creatorcontrib>Sacaleanu, Bogdan</creatorcontrib><creatorcontrib>Gershman, Anatole</creatorcontrib><creatorcontrib>Fano, Andrew</creatorcontrib><creatorcontrib>Hovy, Eduard</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>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>Rajagopal, Dheeraj</au><au>Khetan, Vivek</au><au>Sacaleanu, Bogdan</au><au>Gershman, Anatole</au><au>Fano, Andrew</au><au>Hovy, Eduard</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Template Filling for Controllable Commonsense Reasoning</atitle><jtitle>arXiv.org</jtitle><date>2022-10-14</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study the commonsense reasoning as a template filling task (TemplateCSR) -- where the language models fills reasoning templates with the given constraints as control factors. As an approach to TemplateCSR, we (i) propose a dataset of commonsense reasoning template-expansion pairs and (ii) introduce POTTER, a pretrained sequence-to-sequence model using prompts to perform commonsense reasoning across concepts. Our experiments show that our approach outperforms baselines both in generation metrics and factuality metrics. We also present a detailed error analysis on our approach's ability to reliably perform commonsense reasoning.</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, 2022-10 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2591842709 |
source | Freely Accessible Journals |
subjects | Coders Domains Encoders-Decoders Error analysis Reasoning |
title | Template Filling for Controllable Commonsense Reasoning |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T10%3A36%3A13IST&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=Template%20Filling%20for%20Controllable%20Commonsense%20Reasoning&rft.jtitle=arXiv.org&rft.au=Rajagopal,%20Dheeraj&rft.date=2022-10-14&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2591842709%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2591842709&rft_id=info:pmid/&rfr_iscdi=true |