<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-03770444</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-16T12:08:41+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Are Embedding Spaces Interpretable? Results of an Intrusion Detection Evaluation on a Large French Corpus</title>
            <author role="crp">
              <persName>
                <forename type="first">Thibault</forename>
                <surname>Prouteau</surname>
              </persName>
              <email type="md5">8d05800906d172fbfddf6c1fcdb14b46</email>
              <email type="domain">univ-lemans.fr</email>
              <idno type="idhal" notation="string">thibault-prouteau</idno>
              <idno type="idhal" notation="numeric">743924</idno>
              <idno type="halauthorid" notation="string">50202-743924</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-9610-3191</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=IHcD7JQAAAAJ&amp;hl=fr</idno>
              <idno type="IDREF">https://www.idref.fr/280160879</idno>
              <affiliation ref="#struct-117974"/>
              <affiliation ref="#struct-489552"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Nicolas</forename>
                <surname>Dugué</surname>
              </persName>
              <email type="md5">c2b38de096044c32bcbe23b8a87e61a1</email>
              <email type="domain">univ-lemans.fr</email>
              <idno type="idhal" notation="string">nicolas-dugue</idno>
              <idno type="idhal" notation="numeric">9194</idno>
              <idno type="halauthorid" notation="string">17741-9194</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-6150-1939</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=K4L78K0AAAAJ</idno>
              <idno type="IDREF">https://www.idref.fr/233490280</idno>
              <affiliation ref="#struct-117974"/>
              <affiliation ref="#struct-489552"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Nathalie</forename>
                <surname>Camelin</surname>
              </persName>
              <email type="md5">36a70c18796cb1c39e802c8e6388e8fa</email>
              <email type="domain">univ-lemans.fr</email>
              <idno type="idhal" notation="string">nathalie-camelin</idno>
              <idno type="idhal" notation="numeric">13884</idno>
              <idno type="halauthorid" notation="string">33672-13884</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-6786-1372</idno>
              <idno type="IDREF">https://www.idref.fr/121979318</idno>
              <idno type="GOOGLE SCHOLAR">Dbw0rE0AAAAJ</idno>
              <affiliation ref="#struct-117974"/>
              <affiliation ref="#struct-489552"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Sylvain</forename>
                <surname>Meignier</surname>
              </persName>
              <email type="md5">d092790b429ce58425fafa40d2d41311</email>
              <email type="domain">univ-lemans.fr</email>
              <idno type="idhal" notation="string">sylvain-meignier</idno>
              <idno type="idhal" notation="numeric">11674</idno>
              <idno type="halauthorid" notation="string">33152-11674</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-7687-073X</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=-_DjQVEAAAAJ</idno>
              <idno type="IDREF">https://www.idref.fr/182269086</idno>
              <affiliation ref="#struct-117974"/>
              <affiliation ref="#struct-489552"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Thibault</forename>
                <surname>Prouteau</surname>
              </persName>
              <email type="md5">78c9407b43fbadc0a868edb0c0415b82</email>
              <email type="domain">univ-lemans.fr</email>
            </editor>
            <funder ref="#projanr-56229"/>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2022-09-06 14:22:23</date>
              <date type="whenModified">2026-02-13 15:40:02</date>
              <date type="whenReleased">2022-09-21 11:44:55</date>
              <date type="whenProduced">2022-06-21</date>
              <date type="whenEndEmbargoed">2022-09-06</date>
              <ref type="file" target="https://hal.science/hal-03770444v1/document">
                <date notBefore="2022-09-06"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-03770444v1/file/2022.lrec-1.469.pdf" id="file-3770444-3297031">
                <date notBefore="2022-09-06"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="890541">
                <persName>
                  <forename>Thibault</forename>
                  <surname>Prouteau</surname>
                </persName>
                <email type="md5">78c9407b43fbadc0a868edb0c0415b82</email>
                <email type="domain">univ-lemans.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-03770444</idno>
            <idno type="halUri">https://hal.science/hal-03770444</idno>
            <idno type="halBibtex">prouteau:hal-03770444</idno>
            <idno type="halRefHtml">&lt;i&gt;LREC 2022&lt;/i&gt;, Jun 2022, Marseille, France. pp.4414-4419</idno>
            <idno type="halRef">LREC 2022, Jun 2022, Marseille, France. pp.4414-4419</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-3770444-3297031"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="UNIV-LEMANS">Université du Mans</idno>
            <idno type="stamp" n="LIUM" corresp="UNIV-LEMANS">Laboratoire Informatique de l'Université du Maine</idno>
            <idno type="stamp" n="LIUM-LST" corresp="LIUM">LIUM - Equipe Language and Speech Technology</idno>
            <idno type="stamp" n="ANR">ANR</idno>
            <idno type="stamp" n="ANR-IA-21" corresp="ANR-IA">ANR-IA-21</idno>
            <idno type="stamp" n="ANR-IA">ANR-IA</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Are Embedding Spaces Interpretable? Results of an Intrusion Detection Evaluation on a Large French Corpus</title>
                <author role="crp">
                  <persName>
                    <forename type="first">Thibault</forename>
                    <surname>Prouteau</surname>
                  </persName>
                  <email type="md5">8d05800906d172fbfddf6c1fcdb14b46</email>
                  <email type="domain">univ-lemans.fr</email>
                  <idno type="idhal" notation="string">thibault-prouteau</idno>
                  <idno type="idhal" notation="numeric">743924</idno>
                  <idno type="halauthorid" notation="string">50202-743924</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-9610-3191</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=IHcD7JQAAAAJ&amp;hl=fr</idno>
                  <idno type="IDREF">https://www.idref.fr/280160879</idno>
                  <affiliation ref="#struct-117974"/>
                  <affiliation ref="#struct-489552"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Nicolas</forename>
                    <surname>Dugué</surname>
                  </persName>
                  <email type="md5">c2b38de096044c32bcbe23b8a87e61a1</email>
                  <email type="domain">univ-lemans.fr</email>
                  <idno type="idhal" notation="string">nicolas-dugue</idno>
                  <idno type="idhal" notation="numeric">9194</idno>
                  <idno type="halauthorid" notation="string">17741-9194</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-6150-1939</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=K4L78K0AAAAJ</idno>
                  <idno type="IDREF">https://www.idref.fr/233490280</idno>
                  <affiliation ref="#struct-117974"/>
                  <affiliation ref="#struct-489552"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Nathalie</forename>
                    <surname>Camelin</surname>
                  </persName>
                  <email type="md5">36a70c18796cb1c39e802c8e6388e8fa</email>
                  <email type="domain">univ-lemans.fr</email>
                  <idno type="idhal" notation="string">nathalie-camelin</idno>
                  <idno type="idhal" notation="numeric">13884</idno>
                  <idno type="halauthorid" notation="string">33672-13884</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-6786-1372</idno>
                  <idno type="IDREF">https://www.idref.fr/121979318</idno>
                  <idno type="GOOGLE SCHOLAR">Dbw0rE0AAAAJ</idno>
                  <affiliation ref="#struct-117974"/>
                  <affiliation ref="#struct-489552"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Sylvain</forename>
                    <surname>Meignier</surname>
                  </persName>
                  <email type="md5">d092790b429ce58425fafa40d2d41311</email>
                  <email type="domain">univ-lemans.fr</email>
                  <idno type="idhal" notation="string">sylvain-meignier</idno>
                  <idno type="idhal" notation="numeric">11674</idno>
                  <idno type="halauthorid" notation="string">33152-11674</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-7687-073X</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=-_DjQVEAAAAJ</idno>
                  <idno type="IDREF">https://www.idref.fr/182269086</idno>
                  <affiliation ref="#struct-117974"/>
                  <affiliation ref="#struct-489552"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>LREC 2022</title>
                  <date type="start">2022-06-21</date>
                  <settlement>Marseille</settlement>
                  <country key="FR">France</country>
                </meeting>
                <imprint>
                  <biblScope unit="volume">Proceedings of the Thirteenth Language Resources and Evaluation Conference</biblScope>
                  <biblScope unit="pp">4414-4419</biblScope>
                </imprint>
              </monogr>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">word embeddings</term>
                <term xml:lang="en">interpretability</term>
                <term xml:lang="en">word intrusion task</term>
                <term xml:lang="en">human evaluation</term>
                <term xml:lang="en">french</term>
              </keywords>
              <classCode scheme="halDomain" n="info">Computer Science [cs]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Word embedding methods allow to represent words as vectors in a space that is structured using word co-occurrences so that words with close meanings are close in this space. These vectors are then provided as input to automatic systems to solve natural language processing problems. Because interpretability is a necessary condition to trusting such systems, interpretability of embedding spaces, the first link in the chain is an important issue. In this paper, we thus evaluate the interpretability of vectors extracted with two approaches: SPINE, a k-sparse auto-encoder, and SINr, a graph-based method. This evaluation is based on a Word Intrusion Task with human annotators. It is operated using a large French corpus, and is thus, as far as we know, the first large-scale experiment regarding word embedding interpretability on this language. Furthermore, contrary to the approaches adopted in the literature where the evaluation is performed on a small sample of frequent words, we consider a more realistic use-case where most of the vocabulary is kept for the evaluation. This allows to show how difficult this task is, even though SPINE and SINr show some promising results. In particular, SINr results are obtained with a very low amount of computation compared to SPINE, while being similarly interpretable.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="laboratory" xml:id="struct-117974" status="VALID">
          <idno type="IdRef">035240776</idno>
          <idno type="RNSR">200615275G</idno>
          <idno type="Wikidata">Q51779666</idno>
          <orgName>Laboratoire d'Informatique de l'Université du Mans</orgName>
          <orgName type="acronym">LIUM</orgName>
          <desc>
            <address>
              <addrLine>Avenue Laennec 72085 Le Mans cedex 9</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://lium.univ-lemans.fr/</ref>
          </desc>
          <listRelation>
            <relation name="EA4023" active="#struct-7566" type="direct"/>
          </listRelation>
        </org>
        <org type="researchteam" xml:id="struct-489552" status="VALID">
          <orgName>Equipe Language and Speech Technology</orgName>
          <orgName type="acronym">LST</orgName>
          <desc>
            <address>
              <addrLine>Avenue Laennec 72085 Le Mans cedex 9</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www-lium.univ-lemans.fr</ref>
          </desc>
          <listRelation>
            <relation active="#struct-117974" type="direct"/>
            <relation name="EA4023" active="#struct-7566" type="indirect"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-7566" status="VALID">
          <idno type="ROR">https://ror.org/01mtcc283</idno>
          <orgName>Le Mans Université</orgName>
          <orgName type="acronym">UM</orgName>
          <desc>
            <address>
              <addrLine>Avenue Olivier Messiaen - 72085 Le Mans cedex 9</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.univ-lemans.fr/</ref>
          </desc>
        </org>
      </listOrg>
      <listOrg type="projects">
        <org type="anrProject" xml:id="projanr-56229" status="VALID">
          <idno type="anr">ANR-21-CE23-0010</idno>
          <orgName>DIGING</orgName>
          <desc>Plongements lexicaux temporels et dynamiques basés graphes</desc>
          <date type="start">2021</date>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>