<?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-03100665</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="2026-05-21T09:59:28+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters</title>
            <author role="aut">
              <persName>
                <forename type="first">Hicham</forename>
                <surname>El Boukkouri</surname>
              </persName>
              <email type="md5">36b429789ff40bcc5ff1c0d41719b61e</email>
              <email type="domain">gmail.com</email>
              <idno type="idhal" notation="string">hicham-el-boukkouri</idno>
              <idno type="idhal" notation="numeric">734714</idno>
              <idno type="halauthorid" notation="string">50207-734714</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-1954-1187</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=rK_ER-YAAAAJ</idno>
              <idno type="IDREF">https://www.idref.fr/259277339</idno>
              <affiliation ref="#struct-1041998"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Olivier</forename>
                <surname>Ferret</surname>
              </persName>
              <email type="md5">383cec00a00425ccdb255ad20191a98e</email>
              <email type="domain">cea.fr</email>
              <idno type="idhal" notation="string">olivier-ferret</idno>
              <idno type="idhal" notation="numeric">14770</idno>
              <idno type="halauthorid" notation="string">22591-14770</idno>
              <idno type="IDREF">https://www.idref.fr/155894498</idno>
              <idno type="ORCID">https://orcid.org/0000-0003-0755-2361</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=-mCQhtIAAAAJ</idno>
              <idno type="ARXIV">https://arxiv.org/a/ferret_o_1</idno>
              <affiliation ref="#struct-241319"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Thomas</forename>
                <surname>Lavergne</surname>
              </persName>
              <email type="md5">4537e847a61faa1382bde6c5fa528d54</email>
              <email type="domain">limsi.fr</email>
              <idno type="idhal" notation="string">lavergne-thomas</idno>
              <idno type="idhal" notation="numeric">1296801</idno>
              <idno type="halauthorid" notation="string">513652-1296801</idno>
              <idno type="IDREF">https://www.idref.fr/197749852</idno>
              <idno type="ORCID">https://orcid.org/0009-0009-0338-4648</idno>
              <idno type="VIAF">https://viaf.org/viaf/410148704251536931413</idno>
              <affiliation ref="#struct-1041998"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Hiroshi</forename>
                <surname>Noji</surname>
              </persName>
              <email type="md5">e92c66e4e8aaba7f484b17082e705e3f</email>
              <email type="domain">aist.go.jp</email>
              <idno type="idhal" notation="numeric">1087717</idno>
              <idno type="halauthorid" notation="string">2124393-1087717</idno>
              <affiliation ref="#struct-302425"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Pierre</forename>
                <surname>Zweigenbaum</surname>
              </persName>
              <email type="md5">875a8768fc7f867a01a904fe839de31d</email>
              <email type="domain">limsi.fr</email>
              <idno type="idhal" notation="string">pierre-zweigenbaum</idno>
              <idno type="idhal" notation="numeric">14995</idno>
              <idno type="halauthorid" notation="string">1218-14995</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-8410-4808</idno>
              <idno type="IDREF">https://www.idref.fr/06664268X</idno>
              <affiliation ref="#struct-1041998"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Jun’ichi</forename>
                <surname>Tsujii</surname>
              </persName>
              <email type="md5">e52ebd1df43bc0aeab0fe076d387fb3a</email>
              <email type="domain">microsoft.com</email>
              <idno type="idhal" notation="numeric">1072082</idno>
              <idno type="halauthorid" notation="string">1948176-1072082</idno>
              <affiliation ref="#struct-302425"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Pierre</forename>
                <surname>Zweigenbaum</surname>
              </persName>
              <email type="md5">8d49a886bc6bd6bbb617300dfc935a79</email>
              <email type="domain">lisn.fr</email>
            </editor>
            <funder ref="#projanr-43293"/>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2021-01-06 17:17:58</date>
              <date type="whenModified">2024-09-03 11:16:05</date>
              <date type="whenReleased">2021-01-18 15:23:15</date>
              <date type="whenProduced">2020-12-08</date>
              <date type="whenEndEmbargoed">2021-01-06</date>
              <ref type="file" target="https://hal.science/hal-03100665v1/document">
                <date notBefore="2021-01-06"/>
              </ref>
              <ref type="file" subtype="greenPublisher" n="1" target="https://hal.science/hal-03100665v1/file/ElBoukkouri_COLING2020.pdf" id="file-3100665-2713099">
                <date notBefore="2021-01-06"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="2369">
                <persName>
                  <forename>Pierre</forename>
                  <surname>Zweigenbaum</surname>
                </persName>
                <email type="md5">8d49a886bc6bd6bbb617300dfc935a79</email>
                <email type="domain">lisn.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-03100665</idno>
            <idno type="halUri">https://hal.science/hal-03100665</idno>
            <idno type="halBibtex">elboukkouri:hal-03100665</idno>
            <idno type="halRefHtml">&lt;i&gt;International Conference on Computational Linguistics&lt;/i&gt;, Dec 2020, Barcelona (on line), Spain. pp.6903-6915</idno>
            <idno type="halRef">International Conference on Computational Linguistics, Dec 2020, Barcelona (on line), Spain. pp.6903-6915</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-3100665-2713099"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="CEA">CEA - Commissariat à l'énergie atomique</idno>
            <idno type="stamp" n="CNRS">CNRS - Centre national de la recherche scientifique</idno>
            <idno type="stamp" n="LIMSI">Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur</idno>
            <idno type="stamp" n="DSV" corresp="CEA">Direction de la recherche fondamentale - sciences du vivant</idno>
            <idno type="stamp" n="DRT" corresp="CEA">Direction de la recherche technologique</idno>
            <idno type="stamp" n="CEA-UPSAY" corresp="CEA">CEA - Université Paris-Saclay</idno>
            <idno type="stamp" n="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="CEA-DRF" corresp="CEA">Direction de Recherche Fondamentale</idno>
            <idno type="stamp" n="LIST" corresp="CEA">Laboratoire d'Intégration des Systèmes et des Technologies</idno>
            <idno type="stamp" n="TEST-HALCNRS">Collection test HAL CNRS</idno>
            <idno type="stamp" n="UNIVERSITE-PARIS-SACLAY" corresp="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="ANR">ANR</idno>
            <idno type="stamp" n="LISN">Laboratoire Interdisciplinaire des Sciences du Numérique</idno>
            <idno type="stamp" n="GS-ENGINEERING">Graduate School Sciences de l'Ingénierie et des Systèmes</idno>
            <idno type="stamp" n="GS-COMPUTER-SCIENCE">Graduate School Computer Science</idno>
            <idno type="stamp" n="GS-SPORT-HUMAN-MOVEMENT">Graduate School Sport, Mouvement, Facteurs Humains</idno>
            <idno type="stamp" n="LISN-ILES" corresp="LISN">Information, Langue Ecrite et Signée </idno>
            <idno type="stamp" n="LISN-STL">Sciences et Technologies des Langues</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">CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Hicham</forename>
                    <surname>El Boukkouri</surname>
                  </persName>
                  <email type="md5">36b429789ff40bcc5ff1c0d41719b61e</email>
                  <email type="domain">gmail.com</email>
                  <idno type="idhal" notation="string">hicham-el-boukkouri</idno>
                  <idno type="idhal" notation="numeric">734714</idno>
                  <idno type="halauthorid" notation="string">50207-734714</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-1954-1187</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=rK_ER-YAAAAJ</idno>
                  <idno type="IDREF">https://www.idref.fr/259277339</idno>
                  <affiliation ref="#struct-1041998"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Olivier</forename>
                    <surname>Ferret</surname>
                  </persName>
                  <email type="md5">383cec00a00425ccdb255ad20191a98e</email>
                  <email type="domain">cea.fr</email>
                  <idno type="idhal" notation="string">olivier-ferret</idno>
                  <idno type="idhal" notation="numeric">14770</idno>
                  <idno type="halauthorid" notation="string">22591-14770</idno>
                  <idno type="IDREF">https://www.idref.fr/155894498</idno>
                  <idno type="ORCID">https://orcid.org/0000-0003-0755-2361</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.fr/citations?user=-mCQhtIAAAAJ</idno>
                  <idno type="ARXIV">https://arxiv.org/a/ferret_o_1</idno>
                  <affiliation ref="#struct-241319"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Thomas</forename>
                    <surname>Lavergne</surname>
                  </persName>
                  <email type="md5">4537e847a61faa1382bde6c5fa528d54</email>
                  <email type="domain">limsi.fr</email>
                  <idno type="idhal" notation="string">lavergne-thomas</idno>
                  <idno type="idhal" notation="numeric">1296801</idno>
                  <idno type="halauthorid" notation="string">513652-1296801</idno>
                  <idno type="IDREF">https://www.idref.fr/197749852</idno>
                  <idno type="ORCID">https://orcid.org/0009-0009-0338-4648</idno>
                  <idno type="VIAF">https://viaf.org/viaf/410148704251536931413</idno>
                  <affiliation ref="#struct-1041998"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Hiroshi</forename>
                    <surname>Noji</surname>
                  </persName>
                  <email type="md5">e92c66e4e8aaba7f484b17082e705e3f</email>
                  <email type="domain">aist.go.jp</email>
                  <idno type="idhal" notation="numeric">1087717</idno>
                  <idno type="halauthorid" notation="string">2124393-1087717</idno>
                  <affiliation ref="#struct-302425"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Pierre</forename>
                    <surname>Zweigenbaum</surname>
                  </persName>
                  <email type="md5">875a8768fc7f867a01a904fe839de31d</email>
                  <email type="domain">limsi.fr</email>
                  <idno type="idhal" notation="string">pierre-zweigenbaum</idno>
                  <idno type="idhal" notation="numeric">14995</idno>
                  <idno type="halauthorid" notation="string">1218-14995</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-8410-4808</idno>
                  <idno type="IDREF">https://www.idref.fr/06664268X</idno>
                  <affiliation ref="#struct-1041998"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Jun’ichi</forename>
                    <surname>Tsujii</surname>
                  </persName>
                  <email type="md5">e52ebd1df43bc0aeab0fe076d387fb3a</email>
                  <email type="domain">microsoft.com</email>
                  <idno type="idhal" notation="numeric">1072082</idno>
                  <idno type="halauthorid" notation="string">1948176-1072082</idno>
                  <affiliation ref="#struct-302425"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>International Conference on Computational Linguistics</title>
                  <date type="start">2020-12-08</date>
                  <date type="end">2020-12-13</date>
                  <settlement>Barcelona (on line)</settlement>
                  <country key="ES">Spain</country>
                </meeting>
                <editor>International Committee on Computational Linguistics</editor>
                <imprint>
                  <biblScope unit="serie">Proceedings of the 28th International Conference on Computational Linguistics</biblScope>
                  <biblScope unit="pp">6903-6915</biblScope>
                </imprint>
              </monogr>
              <ref type="publisher">https://coling2020.org/</ref>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <classCode scheme="halDomain" n="info.info-cl">Computer Science [cs]/Computation and Language [cs.CL]</classCode>
              <classCode scheme="halDomain" n="info.info-tt">Computer Science [cs]/Document and Text Processing</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>Due to the compelling improvements brought by BERT, many recent representation models adopted the Transformer architecture as their main building block, consequently inheriting the wordpiece tokenization system despite it not being intrinsically linked to the notion of Transformers. While this system is thought to achieve a good balance between the flexibility of characters and the efficiency of full words, using predefined wordpiece vocabularies from the general domain is not always suitable, especially when building models for specialized domains (e.g., the medical domain). Moreover, adopting a wordpiece tokenization shifts the focus from the word level to the subword level, making the models conceptually more complex and arguably less convenient in practice. For these reasons, we propose CharacterBERT, a new variant of BERT that drops the wordpiece system altogether and uses a Character-CNN module instead to represent entire words by consulting their characters. We show that this new model improves the performance of BERT on a variety of medical domain tasks while at the same time producing robust, word-level, and open-vocabulary representations.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="researchteam" xml:id="struct-1041998" status="OLD">
          <orgName>Information, Langue Ecrite et Signée</orgName>
          <orgName type="acronym">ILES</orgName>
          <date type="start">2020-01-01</date>
          <date type="end">2020-12-31</date>
          <desc>
            <address>
              <addrLine>Bâtiment 508, 91405, Orsay cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.limsi.fr/fr/recherche/iles</ref>
          </desc>
          <listRelation>
            <relation active="#struct-1041968" type="direct"/>
            <relation active="#struct-419361" type="indirect"/>
            <relation name="UPR3251" active="#struct-441569" type="indirect"/>
          </listRelation>
        </org>
        <org type="department" xml:id="struct-241319" status="VALID">
          <idno type="RNSR">200822618E</idno>
          <orgName>Département Intelligence Ambiante et Systèmes Interactifs</orgName>
          <orgName type="acronym">DIASI (CEA, LIST)</orgName>
          <desc>
            <address>
              <addrLine>CEA Saclay - Nanoinnov 91191 Gif-Sur-Yvette Cedex</addrLine>
              <country key="FR"/>
            </address>
          </desc>
          <listRelation>
            <relation name="DRT/LIST/DIASI" active="#struct-40217" type="direct"/>
            <relation name="DRT/LIST" active="#struct-440043" type="indirect"/>
            <relation name="DRT" active="#struct-300016" type="indirect"/>
            <relation active="#struct-419361" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-302425" status="VALID">
          <idno type="ROR">https://ror.org/01703db54</idno>
          <orgName>National Institute of Advanced Industrial Science and Technology</orgName>
          <orgName type="acronym">AIST</orgName>
          <desc>
            <address>
              <addrLine>Higashi, Tsukuba, Ibaraki 305-8561</addrLine>
              <country key="JP"/>
            </address>
            <ref type="url">http://www.aist.go.jp/index_en.html</ref>
          </desc>
        </org>
        <org type="laboratory" xml:id="struct-1041968" status="OLD">
          <orgName>Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur</orgName>
          <orgName type="acronym">LIMSI</orgName>
          <date type="start">2020-01-01</date>
          <date type="end">2020-12-31</date>
          <desc>
            <address>
              <addrLine>Bâtiment 507, rue du Belvédère,91405, Orsay cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.limsi.fr/fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-419361" type="direct"/>
            <relation name="UPR3251" active="#struct-441569" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-419361" status="VALID">
          <idno type="IdRef">241345251</idno>
          <idno type="ROR">https://ror.org/03xjwb503</idno>
          <orgName>Université Paris-Saclay</orgName>
          <desc>
            <address>
              <addrLine>Bâtiment Bréguet, 3 Rue Joliot Curie 2e ét, 91190 Gif-sur-Yvette</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.universite-paris-saclay.fr/fr</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-441569" status="VALID">
          <idno type="IdRef">02636817X</idno>
          <idno type="ISNI">0000000122597504</idno>
          <idno type="ROR">https://ror.org/02feahw73</idno>
          <orgName>Centre National de la Recherche Scientifique</orgName>
          <orgName type="acronym">CNRS</orgName>
          <date type="start">1939-10-19</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.cnrs.fr/</ref>
          </desc>
        </org>
        <org type="laboratory" xml:id="struct-40217" status="VALID">
          <idno type="IdRef">156836882</idno>
          <idno type="ISNI">0000 0004 0405 1788</idno>
          <idno type="RNSR">200118591H</idno>
          <idno type="ROR">https://ror.org/000dbcc61</idno>
          <idno type="Wikidata">Q30299467</idno>
          <orgName>Laboratoire d'Intégration des Systèmes et des Technologies</orgName>
          <orgName type="acronym">LIST (CEA)</orgName>
          <desc>
            <address>
              <addrLine>DRT/LISTNano-INNOVAvenue de la Vauve91120 Palaiseau</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www-list.cea.fr/</ref>
          </desc>
          <listRelation>
            <relation name="DRT/LIST" active="#struct-440043" type="direct"/>
            <relation name="DRT" active="#struct-300016" type="indirect"/>
          </listRelation>
        </org>
        <org type="regrouplaboratory" xml:id="struct-440043" status="VALID">
          <idno type="IdRef">067087930</idno>
          <idno type="ISNI">0000000121157881</idno>
          <idno type="RNSR">199018589D</idno>
          <idno type="ROR">https://ror.org/02ggzyd20</idno>
          <idno type="Wikidata">Q30299418</idno>
          <orgName>Direction de Recherche Technologique (CEA)</orgName>
          <orgName type="acronym">DRT (CEA)</orgName>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.drt-cea.com/</ref>
          </desc>
          <listRelation>
            <relation name="DRT" active="#struct-300016" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-300016" status="VALID">
          <idno type="IdRef">026372061</idno>
          <idno type="ISNI">0000000122998025</idno>
          <idno type="ROR">https://ror.org/00jjx8s55</idno>
          <idno type="Wikidata">Q868550</idno>
          <orgName>Commissariat à l'énergie atomique et aux énergies alternatives</orgName>
          <orgName type="acronym">CEA</orgName>
          <desc>
            <address>
              <addrLine>Centre de SaclayCentre de GrenobleCentre de Cadaracheetc</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.cea.fr/</ref>
          </desc>
        </org>
      </listOrg>
      <listOrg type="projects">
        <org type="anrProject" xml:id="projanr-43293" status="VALID">
          <idno type="anr">ANR-17-CE23-0001</idno>
          <orgName>ADDICTE</orgName>
          <desc>Analyse distributionnelle en domaine de spécialité</desc>
          <date type="start">2017</date>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>