Neuro-Symbolic Visual Dialog
We propose Neuro-Symbolic Visual Dialog (NSVD) -the first method to combine deep learning and symbolic program execution for multi-round visually-grounded reasoning. NSVD significantly outperforms existing purely-connectionist methods on two key challenges inherent to visual dialog: long-distance co...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2022-08 |
---|---|
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 | Abdessaied, Adnen Bâce, Mihai Bulling, Andreas |
description | We propose Neuro-Symbolic Visual Dialog (NSVD) -the first method to combine deep learning and symbolic program execution for multi-round visually-grounded reasoning. NSVD significantly outperforms existing purely-connectionist methods on two key challenges inherent to visual dialog: long-distance co-reference resolution as well as vanishing question-answering performance. We demonstrate the latter by proposing a more realistic and stricter evaluation scheme in which we use predicted answers for the full dialog history when calculating accuracy. We describe two variants of our model and show that using this new scheme, our best model achieves an accuracy of 99.72% on CLEVR-Dialog -a relative improvement of more than 10% over the state of the art while only requiring a fraction of training data. Moreover, we demonstrate that our neuro-symbolic models have a higher mean first failure round, are more robust against incomplete dialog histories, and generalise better not only to dialogs that are up to three times longer than those seen during training but also to unseen question types and scenes. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2705552788</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2705552788</sourcerecordid><originalsourceid>FETCH-proquest_journals_27055527883</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSQ8UstLcrXDa7MTcrPyUxWCMssLk3MUXDJTMzJT-dhYE1LzClO5YXS3AzKbq4hzh66BUX5haWpxSXxWfmlRXlAqXgjcwNTU1MjcwsLY-JUAQAbOit_</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2705552788</pqid></control><display><type>article</type><title>Neuro-Symbolic Visual Dialog</title><source>Free E- Journals</source><creator>Abdessaied, Adnen ; Bâce, Mihai ; Bulling, Andreas</creator><creatorcontrib>Abdessaied, Adnen ; Bâce, Mihai ; Bulling, Andreas</creatorcontrib><description>We propose Neuro-Symbolic Visual Dialog (NSVD) -the first method to combine deep learning and symbolic program execution for multi-round visually-grounded reasoning. NSVD significantly outperforms existing purely-connectionist methods on two key challenges inherent to visual dialog: long-distance co-reference resolution as well as vanishing question-answering performance. We demonstrate the latter by proposing a more realistic and stricter evaluation scheme in which we use predicted answers for the full dialog history when calculating accuracy. We describe two variants of our model and show that using this new scheme, our best model achieves an accuracy of 99.72% on CLEVR-Dialog -a relative improvement of more than 10% over the state of the art while only requiring a fraction of training data. Moreover, we demonstrate that our neuro-symbolic models have a higher mean first failure round, are more robust against incomplete dialog histories, and generalise better not only to dialogs that are up to three times longer than those seen during training but also to unseen question types and scenes.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Model accuracy ; Questions ; Training</subject><ispartof>arXiv.org, 2022-08</ispartof><rights>2022. This work is published under http://creativecommons.org/licenses/by-nc-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><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Abdessaied, Adnen</creatorcontrib><creatorcontrib>Bâce, Mihai</creatorcontrib><creatorcontrib>Bulling, Andreas</creatorcontrib><title>Neuro-Symbolic Visual Dialog</title><title>arXiv.org</title><description>We propose Neuro-Symbolic Visual Dialog (NSVD) -the first method to combine deep learning and symbolic program execution for multi-round visually-grounded reasoning. NSVD significantly outperforms existing purely-connectionist methods on two key challenges inherent to visual dialog: long-distance co-reference resolution as well as vanishing question-answering performance. We demonstrate the latter by proposing a more realistic and stricter evaluation scheme in which we use predicted answers for the full dialog history when calculating accuracy. We describe two variants of our model and show that using this new scheme, our best model achieves an accuracy of 99.72% on CLEVR-Dialog -a relative improvement of more than 10% over the state of the art while only requiring a fraction of training data. Moreover, we demonstrate that our neuro-symbolic models have a higher mean first failure round, are more robust against incomplete dialog histories, and generalise better not only to dialogs that are up to three times longer than those seen during training but also to unseen question types and scenes.</description><subject>Model accuracy</subject><subject>Questions</subject><subject>Training</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSQ8UstLcrXDa7MTcrPyUxWCMssLk3MUXDJTMzJT-dhYE1LzClO5YXS3AzKbq4hzh66BUX5haWpxSXxWfmlRXlAqXgjcwNTU1MjcwsLY-JUAQAbOit_</recordid><startdate>20220822</startdate><enddate>20220822</enddate><creator>Abdessaied, Adnen</creator><creator>Bâce, Mihai</creator><creator>Bulling, Andreas</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>20220822</creationdate><title>Neuro-Symbolic Visual Dialog</title><author>Abdessaied, Adnen ; Bâce, Mihai ; Bulling, Andreas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27055527883</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Model accuracy</topic><topic>Questions</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Abdessaied, Adnen</creatorcontrib><creatorcontrib>Bâce, Mihai</creatorcontrib><creatorcontrib>Bulling, Andreas</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>Abdessaied, Adnen</au><au>Bâce, Mihai</au><au>Bulling, Andreas</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Neuro-Symbolic Visual Dialog</atitle><jtitle>arXiv.org</jtitle><date>2022-08-22</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>We propose Neuro-Symbolic Visual Dialog (NSVD) -the first method to combine deep learning and symbolic program execution for multi-round visually-grounded reasoning. NSVD significantly outperforms existing purely-connectionist methods on two key challenges inherent to visual dialog: long-distance co-reference resolution as well as vanishing question-answering performance. We demonstrate the latter by proposing a more realistic and stricter evaluation scheme in which we use predicted answers for the full dialog history when calculating accuracy. We describe two variants of our model and show that using this new scheme, our best model achieves an accuracy of 99.72% on CLEVR-Dialog -a relative improvement of more than 10% over the state of the art while only requiring a fraction of training data. Moreover, we demonstrate that our neuro-symbolic models have a higher mean first failure round, are more robust against incomplete dialog histories, and generalise better not only to dialogs that are up to three times longer than those seen during training but also to unseen question types and scenes.</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, 2022-08 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2705552788 |
source | Free E- Journals |
subjects | Model accuracy Questions Training |
title | Neuro-Symbolic Visual Dialog |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T10%3A27%3A06IST&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=Neuro-Symbolic%20Visual%20Dialog&rft.jtitle=arXiv.org&rft.au=Abdessaied,%20Adnen&rft.date=2022-08-22&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2705552788%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2705552788&rft_id=info:pmid/&rfr_iscdi=true |