No Cover Image

Conference Paper/Proceeding/Abstract 242 views 32 downloads

Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues

Julian Hough Orcid Logo, Sina Zarrieß, Casey Kennington, David Schlangen, Massimo Poesio

The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation, Pages: 3689 - 3699

Swansea University Author: Julian Hough Orcid Logo

  • 65924.VoR.pdf

    PDF | Version of Record

    © 2024 ELRA Language Resource Association: CC BY-NC 4.0

    Download (2.39MB)

Check full text

Abstract

Using Brennan and Clark’s theory of a Conceptual Pact, that when interlocutors agree on a name for an object, they are forming a temporary agreement on how to conceptualize that object, we present an extension to a simple reference resolver which simulates this process over time with different conve...

Full description

Published in: The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation
ISBN: 9782493814104
ISSN: 2522-2686
Published: ELRA and ICCL 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa65924
first_indexed 2024-03-27T16:18:27Z
last_indexed 2024-11-25T14:17:07Z
id cronfa65924
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2024-06-27T14:04:51.6799233</datestamp><bib-version>v2</bib-version><id>65924</id><entry>2024-03-27</entry><title>Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues</title><swanseaauthors><author><sid>082d773ae261d2bbf49434dd2608ab40</sid><ORCID>0000-0002-4345-6759</ORCID><firstname>Julian</firstname><surname>Hough</surname><name>Julian Hough</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-03-27</date><deptcode>MACS</deptcode><abstract>Using Brennan and Clark&#x2019;s theory of a Conceptual Pact, that when interlocutors agree on a name for an object, they are forming a temporary agreement on how to conceptualize that object, we present an extension to a simple reference resolver which simulates this process over time with different conversation pairs. In a puzzle construction domain, we model pacts with small language models for each referent which update during the interaction. When features from these pact models are incorporated into a simple bag-of-words reference resolver, the accuracy increases compared to using a standard pre-trained model. The model performs equally to a competitor using the same data but with exhaustive re-training after each prediction, while also being more transparent, faster and less resource-intensive. We also experiment with reducing the number of training interactions, and can still achieve reference resolution accuracies of over 80% in testing from observing a single previous interaction, over 20% higher than a pre-trained baseline. While this is a limited domain, we argue the model could be applicable to larger real-world applications in human and human-robot interaction and is an interpretable and transparent model.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation</journal><volume/><journalNumber/><paginationStart>3689</paginationStart><paginationEnd>3699</paginationEnd><publisher>ELRA and ICCL</publisher><placeOfPublication/><isbnPrint/><isbnElectronic>9782493814104</isbnElectronic><issnPrint/><issnElectronic>2522-2686</issnElectronic><keywords/><publishedDay>20</publishedDay><publishedMonth>5</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-05-20</publishedDate><doi/><url>https://aclanthology.org/2024.lrec-main.327/</url><notes>https://aclanthology.org/2024.lrec-main.327/</notes><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm>Not Required</apcterm><funders>Hough is supported by the UKRI Engineering and Physical Sciences Research Council (EPSRC) grant EP/X009343/1 &#x2018;FLUIDITY&#x2019; and Poesio and Hough are supported by EPSRC grant EP/W001632/1 &#x2018;ARCIDUCA&#x2019;.</funders><projectreference/><lastEdited>2024-06-27T14:04:51.6799233</lastEdited><Created>2024-03-27T16:12:34.5119314</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Julian</firstname><surname>Hough</surname><orcid>0000-0002-4345-6759</orcid><order>1</order></author><author><firstname>Sina</firstname><surname>Zarrie&#xDF;</surname><order>2</order></author><author><firstname>Casey</firstname><surname>Kennington</surname><order>3</order></author><author><firstname>David</firstname><surname>Schlangen</surname><order>4</order></author><author><firstname>Massimo</firstname><surname>Poesio</surname><order>5</order></author></authors><documents><document><filename>65924__30768__3547826b9b18458bb5de9314d8bd5a92.pdf</filename><originalFilename>65924.VoR.pdf</originalFilename><uploaded>2024-06-27T13:49:05.9502107</uploaded><type>Output</type><contentLength>2509426</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2024 ELRA Language Resource Association: CC BY-NC 4.0</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by-nc-sa/3.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2024-06-27T14:04:51.6799233 v2 65924 2024-03-27 Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues 082d773ae261d2bbf49434dd2608ab40 0000-0002-4345-6759 Julian Hough Julian Hough true false 2024-03-27 MACS Using Brennan and Clark’s theory of a Conceptual Pact, that when interlocutors agree on a name for an object, they are forming a temporary agreement on how to conceptualize that object, we present an extension to a simple reference resolver which simulates this process over time with different conversation pairs. In a puzzle construction domain, we model pacts with small language models for each referent which update during the interaction. When features from these pact models are incorporated into a simple bag-of-words reference resolver, the accuracy increases compared to using a standard pre-trained model. The model performs equally to a competitor using the same data but with exhaustive re-training after each prediction, while also being more transparent, faster and less resource-intensive. We also experiment with reducing the number of training interactions, and can still achieve reference resolution accuracies of over 80% in testing from observing a single previous interaction, over 20% higher than a pre-trained baseline. While this is a limited domain, we argue the model could be applicable to larger real-world applications in human and human-robot interaction and is an interpretable and transparent model. Conference Paper/Proceeding/Abstract The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation 3689 3699 ELRA and ICCL 9782493814104 2522-2686 20 5 2024 2024-05-20 https://aclanthology.org/2024.lrec-main.327/ https://aclanthology.org/2024.lrec-main.327/ COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University Not Required Hough is supported by the UKRI Engineering and Physical Sciences Research Council (EPSRC) grant EP/X009343/1 ‘FLUIDITY’ and Poesio and Hough are supported by EPSRC grant EP/W001632/1 ‘ARCIDUCA’. 2024-06-27T14:04:51.6799233 2024-03-27T16:12:34.5119314 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Julian Hough 0000-0002-4345-6759 1 Sina Zarrieß 2 Casey Kennington 3 David Schlangen 4 Massimo Poesio 5 65924__30768__3547826b9b18458bb5de9314d8bd5a92.pdf 65924.VoR.pdf 2024-06-27T13:49:05.9502107 Output 2509426 application/pdf Version of Record true © 2024 ELRA Language Resource Association: CC BY-NC 4.0 true eng https://creativecommons.org/licenses/by-nc-sa/3.0/
title Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
spellingShingle Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
Julian Hough
title_short Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
title_full Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
title_fullStr Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
title_full_unstemmed Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
title_sort Conceptual Pacts for Reference Resolution using Small, Dynamically Constructed Language Models: A Study in Puzzle Building Dialogues
author_id_str_mv 082d773ae261d2bbf49434dd2608ab40
author_id_fullname_str_mv 082d773ae261d2bbf49434dd2608ab40_***_Julian Hough
author Julian Hough
author2 Julian Hough
Sina Zarrieß
Casey Kennington
David Schlangen
Massimo Poesio
format Conference Paper/Proceeding/Abstract
container_title The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation
container_start_page 3689
publishDate 2024
institution Swansea University
isbn 9782493814104
issn 2522-2686
publisher ELRA and ICCL
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
url https://aclanthology.org/2024.lrec-main.327/
document_store_str 1
active_str 0
description Using Brennan and Clark’s theory of a Conceptual Pact, that when interlocutors agree on a name for an object, they are forming a temporary agreement on how to conceptualize that object, we present an extension to a simple reference resolver which simulates this process over time with different conversation pairs. In a puzzle construction domain, we model pacts with small language models for each referent which update during the interaction. When features from these pact models are incorporated into a simple bag-of-words reference resolver, the accuracy increases compared to using a standard pre-trained model. The model performs equally to a competitor using the same data but with exhaustive re-training after each prediction, while also being more transparent, faster and less resource-intensive. We also experiment with reducing the number of training interactions, and can still achieve reference resolution accuracies of over 80% in testing from observing a single previous interaction, over 20% higher than a pre-trained baseline. While this is a limited domain, we argue the model could be applicable to larger real-world applications in human and human-robot interaction and is an interpretable and transparent model.
published_date 2024-05-20T20:29:22Z
_version_ 1821348160944996352
score 11.04748

Similar Items