SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation

SimpleMTOD is a simple language model which recasts several sub-tasks in multimodal task-oriented dialogues as sequence prediction tasks. SimpleMTOD is built on a large-scale transformer-based auto-regressive architecture, which has already proven to be successful in uni-modal task-oriented dialogue...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hemanthage, Bhathiya, Dondrup, Christian, Bartie, Phil, Lemon, Oliver
Format: Artikel
Sprache:eng
Schlagworte:
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 Hemanthage, Bhathiya
Dondrup, Christian
Bartie, Phil
Lemon, Oliver
description SimpleMTOD is a simple language model which recasts several sub-tasks in multimodal task-oriented dialogues as sequence prediction tasks. SimpleMTOD is built on a large-scale transformer-based auto-regressive architecture, which has already proven to be successful in uni-modal task-oriented dialogues, and effectively leverages transfer learning from pre-trained GPT-2. In-order to capture the semantics of visual scenes, we introduce both local and de-localized tokens for objects within a scene. De-localized tokens represent the type of an object rather than the specific object itself and so possess a consistent meaning across the dataset. SimpleMTOD achieves a state-of-the-art BLEU score (0.327) in the Response Generation sub-task of the SIMMC 2.0 test-std dataset while performing on par in other multimodal sub-tasks: Disambiguation, Coreference Resolution, and Dialog State Tracking. This is despite taking a minimalist approach for extracting visual (and non-visual) information. In addition the model does not rely on task-specific architectural changes such as classification heads.
doi_str_mv 10.48550/arxiv.2307.04907
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2307_04907</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2307_04907</sourcerecordid><originalsourceid>FETCH-LOGICAL-a677-c44a5a2c09b9d4d80687d74b8f2a17d82d86377e99a258480dbc0a3eff25b7c13</originalsourceid><addsrcrecordid>eNotz8tOhDAYBeBuXJjRB3BlXwAspdDibjLjLWFCIuzJT_uDjYUSLuq8vc6Mq5OTnJzkI-QuYqFQScIeYPqxXyGPmQyZyJi8Jq60_ejwUBX7R7qll0ZzGLoVOqQHb9DR1k_0sLrF9t6AoxXMn0ExWRwWNHRvwfluRfptlw9aHvvGO6tpqXFA-o7jhPPfEBbrhxty1YKb8fY_N6R6fqp2r0FevLzttnkAqZSBFgIS4JplTWaEUSxV0kjRqJZDJI3iRqWxlJhlwBMlFDONZhBj2_KkkTqKN-T-cnvm1uNke5iO9Yldn9nxL7U5U2I</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation</title><source>arXiv.org</source><creator>Hemanthage, Bhathiya ; Dondrup, Christian ; Bartie, Phil ; Lemon, Oliver</creator><creatorcontrib>Hemanthage, Bhathiya ; Dondrup, Christian ; Bartie, Phil ; Lemon, Oliver</creatorcontrib><description>SimpleMTOD is a simple language model which recasts several sub-tasks in multimodal task-oriented dialogues as sequence prediction tasks. SimpleMTOD is built on a large-scale transformer-based auto-regressive architecture, which has already proven to be successful in uni-modal task-oriented dialogues, and effectively leverages transfer learning from pre-trained GPT-2. In-order to capture the semantics of visual scenes, we introduce both local and de-localized tokens for objects within a scene. De-localized tokens represent the type of an object rather than the specific object itself and so possess a consistent meaning across the dataset. SimpleMTOD achieves a state-of-the-art BLEU score (0.327) in the Response Generation sub-task of the SIMMC 2.0 test-std dataset while performing on par in other multimodal sub-tasks: Disambiguation, Coreference Resolution, and Dialog State Tracking. This is despite taking a minimalist approach for extracting visual (and non-visual) information. In addition the model does not rely on task-specific architectural changes such as classification heads.</description><identifier>DOI: 10.48550/arxiv.2307.04907</identifier><language>eng</language><subject>Computer Science - Computation and Language ; Computer Science - Learning</subject><creationdate>2023-07</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2307.04907$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2307.04907$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Hemanthage, Bhathiya</creatorcontrib><creatorcontrib>Dondrup, Christian</creatorcontrib><creatorcontrib>Bartie, Phil</creatorcontrib><creatorcontrib>Lemon, Oliver</creatorcontrib><title>SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation</title><description>SimpleMTOD is a simple language model which recasts several sub-tasks in multimodal task-oriented dialogues as sequence prediction tasks. SimpleMTOD is built on a large-scale transformer-based auto-regressive architecture, which has already proven to be successful in uni-modal task-oriented dialogues, and effectively leverages transfer learning from pre-trained GPT-2. In-order to capture the semantics of visual scenes, we introduce both local and de-localized tokens for objects within a scene. De-localized tokens represent the type of an object rather than the specific object itself and so possess a consistent meaning across the dataset. SimpleMTOD achieves a state-of-the-art BLEU score (0.327) in the Response Generation sub-task of the SIMMC 2.0 test-std dataset while performing on par in other multimodal sub-tasks: Disambiguation, Coreference Resolution, and Dialog State Tracking. This is despite taking a minimalist approach for extracting visual (and non-visual) information. In addition the model does not rely on task-specific architectural changes such as classification heads.</description><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz8tOhDAYBeBuXJjRB3BlXwAspdDibjLjLWFCIuzJT_uDjYUSLuq8vc6Mq5OTnJzkI-QuYqFQScIeYPqxXyGPmQyZyJi8Jq60_ejwUBX7R7qll0ZzGLoVOqQHb9DR1k_0sLrF9t6AoxXMn0ExWRwWNHRvwfluRfptlw9aHvvGO6tpqXFA-o7jhPPfEBbrhxty1YKb8fY_N6R6fqp2r0FevLzttnkAqZSBFgIS4JplTWaEUSxV0kjRqJZDJI3iRqWxlJhlwBMlFDONZhBj2_KkkTqKN-T-cnvm1uNke5iO9Yldn9nxL7U5U2I</recordid><startdate>20230710</startdate><enddate>20230710</enddate><creator>Hemanthage, Bhathiya</creator><creator>Dondrup, Christian</creator><creator>Bartie, Phil</creator><creator>Lemon, Oliver</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230710</creationdate><title>SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation</title><author>Hemanthage, Bhathiya ; Dondrup, Christian ; Bartie, Phil ; Lemon, Oliver</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a677-c44a5a2c09b9d4d80687d74b8f2a17d82d86377e99a258480dbc0a3eff25b7c13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Hemanthage, Bhathiya</creatorcontrib><creatorcontrib>Dondrup, Christian</creatorcontrib><creatorcontrib>Bartie, Phil</creatorcontrib><creatorcontrib>Lemon, Oliver</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Hemanthage, Bhathiya</au><au>Dondrup, Christian</au><au>Bartie, Phil</au><au>Lemon, Oliver</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation</atitle><date>2023-07-10</date><risdate>2023</risdate><abstract>SimpleMTOD is a simple language model which recasts several sub-tasks in multimodal task-oriented dialogues as sequence prediction tasks. SimpleMTOD is built on a large-scale transformer-based auto-regressive architecture, which has already proven to be successful in uni-modal task-oriented dialogues, and effectively leverages transfer learning from pre-trained GPT-2. In-order to capture the semantics of visual scenes, we introduce both local and de-localized tokens for objects within a scene. De-localized tokens represent the type of an object rather than the specific object itself and so possess a consistent meaning across the dataset. SimpleMTOD achieves a state-of-the-art BLEU score (0.327) in the Response Generation sub-task of the SIMMC 2.0 test-std dataset while performing on par in other multimodal sub-tasks: Disambiguation, Coreference Resolution, and Dialog State Tracking. This is despite taking a minimalist approach for extracting visual (and non-visual) information. In addition the model does not rely on task-specific architectural changes such as classification heads.</abstract><doi>10.48550/arxiv.2307.04907</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2307.04907
ispartof
issn
language eng
recordid cdi_arxiv_primary_2307_04907
source arXiv.org
subjects Computer Science - Computation and Language
Computer Science - Learning
title SimpleMTOD: A Simple Language Model for Multimodal Task-Oriented Dialogue with Symbolic Scene Representation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-31T12%3A51%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=SimpleMTOD:%20A%20Simple%20Language%20Model%20for%20Multimodal%20Task-Oriented%20Dialogue%20with%20Symbolic%20Scene%20Representation&rft.au=Hemanthage,%20Bhathiya&rft.date=2023-07-10&rft_id=info:doi/10.48550/arxiv.2307.04907&rft_dat=%3Carxiv_GOX%3E2307_04907%3C/arxiv_GOX%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