<?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-05572941</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="2026-05-03T10:19:30+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">LLM-based Atomic Propositions Help Weak Extractors: Evaluation of a Propositioner for Triplet Extraction</title>
            <author role="aut">
              <persName>
                <forename type="first">Luc</forename>
                <surname>Pommeret</surname>
              </persName>
              <email type="md5">f30dc54d51effd2d7c784eb596c82ee4</email>
              <email type="domain">universite-paris-saclay.fr</email>
              <ptr type="url" target="https://lucpommeret.com"/>
              <idno type="idhal" notation="string">pommeret</idno>
              <idno type="idhal" notation="numeric">1535712</idno>
              <idno type="halauthorid" notation="string">3495804-1535712</idno>
              <idno type="GOOGLE SCHOLAR">f-6PeWUAAAAJ</idno>
              <affiliation ref="#struct-1065372"/>
              <affiliation ref="#struct-419361"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Thomas</forename>
                <surname>Gerald</surname>
              </persName>
              <email type="md5">02d454afceebf87b1da11ac7f7722f60</email>
              <email type="domain">universite-paris-saclay.fr</email>
              <idno type="idhal" notation="string">thomasgerald</idno>
              <idno type="idhal" notation="numeric">1217894</idno>
              <idno type="halauthorid" notation="string">1821181-1217894</idno>
              <idno type="ORCID">https://orcid.org/0000-0001-7510-9690</idno>
              <idno type="IDREF">https://www.idref.fr/258617519</idno>
              <idno type="VIAF">https://viaf.org/viaf/245163936564300832150</idno>
              <affiliation ref="#struct-1061259"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Patrick</forename>
                <surname>Paroubek</surname>
              </persName>
              <email type="md5">7e7397fc00a8cd40050afba779c85168</email>
              <email type="domain">limsi.fr</email>
              <idno type="idhal" notation="string">patrick-paroubek</idno>
              <idno type="idhal" notation="numeric">20704</idno>
              <idno type="halauthorid" notation="string">3077-20704</idno>
              <idno type="IDREF">https://www.idref.fr/057218730</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-4302-1894</idno>
              <idno type="ISNI">http://isni.org/isni/0000000432402557</idno>
              <idno type="VIAF">https://viaf.org/viaf/214857358</idno>
              <affiliation ref="#struct-1065372"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Sahar</forename>
                <surname>Ghannay</surname>
              </persName>
              <email type="md5">17b7bf53bc631ff63c3c8f0a4057ebba</email>
              <email type="domain">gmail.com</email>
              <idno type="idhal" notation="string">sahar-ghannay</idno>
              <idno type="idhal" notation="numeric">16545</idno>
              <idno type="halauthorid" notation="string">33729-16545</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-7531-2522</idno>
              <idno type="IDREF">https://www.idref.fr/220420521</idno>
              <idno type="VIAF">https://viaf.org/viaf/119151302887948660077</idno>
              <affiliation ref="#struct-1065372"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Christophe</forename>
                <surname>Servan</surname>
              </persName>
              <email type="md5">65a2a8d7160b6e43a6e9caa24d9af1f2</email>
              <email type="domain">gmail.com</email>
              <idno type="idhal" notation="string">servan</idno>
              <idno type="idhal" notation="numeric">4128</idno>
              <idno type="halauthorid" notation="string">9504-4128</idno>
              <idno type="IDREF">https://www.idref.fr/147053609</idno>
              <idno type="ORCID">https://orcid.org/0000-0003-2306-7075</idno>
              <affiliation ref="#struct-1065372"/>
              <affiliation ref="#struct-1222886"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Sophie</forename>
                <surname>Rosset</surname>
              </persName>
              <email type="md5">dbb4013fd09f2b07c31e3a8e3878f7bc</email>
              <email type="domain">limsi.fr</email>
              <idno type="idhal" notation="string">sophie-rosset</idno>
              <idno type="idhal" notation="numeric">14913</idno>
              <idno type="halauthorid" notation="string">11264-14913</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-6865-4989</idno>
              <idno type="IDREF">https://www.idref.fr/137157835</idno>
              <idno type="ISNI">http://isni.org/isni/0000000357034976</idno>
              <idno type="VIAF">https://viaf.org/viaf/193001246</idno>
              <affiliation ref="#struct-1061259"/>
              <affiliation ref="#struct-1065372"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Luc</forename>
                <surname>Pommeret</surname>
              </persName>
              <email type="md5">f30dc54d51effd2d7c784eb596c82ee4</email>
              <email type="domain">universite-paris-saclay.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2026-04-02 22:36:55</date>
              <date type="whenModified">2026-04-06 12:43:08</date>
              <date type="whenReleased">2026-04-03 10:31:46</date>
              <date type="whenProduced">2026-05-16</date>
              <date type="whenEndEmbargoed">2026-04-02</date>
              <ref type="file" target="https://hal.science/hal-05572941v1/document">
                <date notBefore="2026-04-02"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-05572941v1/file/ARTICLE_PROPRE.pdf" id="file-5572941-4777659">
                <date notBefore="2026-04-02"/>
              </ref>
              <ref type="externalLink" target="http://arxiv.org/pdf/2604.02866"/>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="1667132">
                <persName>
                  <forename>Luc</forename>
                  <surname>Pommeret</surname>
                </persName>
                <email type="md5">f30dc54d51effd2d7c784eb596c82ee4</email>
                <email type="domain">universite-paris-saclay.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-05572941</idno>
            <idno type="halUri">https://hal.science/hal-05572941</idno>
            <idno type="halBibtex">pommeret:hal-05572941</idno>
            <idno type="halRefHtml">&lt;i&gt;KG-LLM@LREC - Knowledge Graphs and Large Language Models&lt;/i&gt;, ELRA, May 2026, Palma De Majorque, Spain</idno>
            <idno type="halRef">KG-LLM@LREC - Knowledge Graphs and Large Language Models, ELRA, May 2026, Palma De Majorque, Spain</idno>
            <availability status="restricted">
              <licence target="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0 - Attribution<ref corresp="#file-5572941-4777659"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="CNRS">CNRS - Centre national de la recherche scientifique</idno>
            <idno type="stamp" n="CENTRALESUPELEC">Ecole CentraleSupélec</idno>
            <idno type="stamp" n="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="UNIVERSITE-PARIS-SACLAY" corresp="UNIV-PARIS-SACLAY">Université Paris-Saclay</idno>
            <idno type="stamp" n="LISN">Laboratoire Interdisciplinaire des Sciences du Numérique</idno>
            <idno type="stamp" n="GS-COMPUTER-SCIENCE">Graduate School Computer Science</idno>
            <idno type="stamp" n="LISN-STL">Sciences et Technologies des Langues</idno>
            <idno type="stamp" n="AMIAD">Agence Ministérielle pour l’Intelligence Artificielle de Défense</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="0">No</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">LLM-based Atomic Propositions Help Weak Extractors: Evaluation of a Propositioner for Triplet Extraction</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Luc</forename>
                    <surname>Pommeret</surname>
                  </persName>
                  <email type="md5">f30dc54d51effd2d7c784eb596c82ee4</email>
                  <email type="domain">universite-paris-saclay.fr</email>
                  <ptr type="url" target="https://lucpommeret.com"/>
                  <idno type="idhal" notation="string">pommeret</idno>
                  <idno type="idhal" notation="numeric">1535712</idno>
                  <idno type="halauthorid" notation="string">3495804-1535712</idno>
                  <idno type="GOOGLE SCHOLAR">f-6PeWUAAAAJ</idno>
                  <affiliation ref="#struct-1065372"/>
                  <affiliation ref="#struct-419361"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Thomas</forename>
                    <surname>Gerald</surname>
                  </persName>
                  <email type="md5">02d454afceebf87b1da11ac7f7722f60</email>
                  <email type="domain">universite-paris-saclay.fr</email>
                  <idno type="idhal" notation="string">thomasgerald</idno>
                  <idno type="idhal" notation="numeric">1217894</idno>
                  <idno type="halauthorid" notation="string">1821181-1217894</idno>
                  <idno type="ORCID">https://orcid.org/0000-0001-7510-9690</idno>
                  <idno type="IDREF">https://www.idref.fr/258617519</idno>
                  <idno type="VIAF">https://viaf.org/viaf/245163936564300832150</idno>
                  <affiliation ref="#struct-1061259"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Patrick</forename>
                    <surname>Paroubek</surname>
                  </persName>
                  <email type="md5">7e7397fc00a8cd40050afba779c85168</email>
                  <email type="domain">limsi.fr</email>
                  <idno type="idhal" notation="string">patrick-paroubek</idno>
                  <idno type="idhal" notation="numeric">20704</idno>
                  <idno type="halauthorid" notation="string">3077-20704</idno>
                  <idno type="IDREF">https://www.idref.fr/057218730</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-4302-1894</idno>
                  <idno type="ISNI">http://isni.org/isni/0000000432402557</idno>
                  <idno type="VIAF">https://viaf.org/viaf/214857358</idno>
                  <affiliation ref="#struct-1065372"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Sahar</forename>
                    <surname>Ghannay</surname>
                  </persName>
                  <email type="md5">17b7bf53bc631ff63c3c8f0a4057ebba</email>
                  <email type="domain">gmail.com</email>
                  <idno type="idhal" notation="string">sahar-ghannay</idno>
                  <idno type="idhal" notation="numeric">16545</idno>
                  <idno type="halauthorid" notation="string">33729-16545</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-7531-2522</idno>
                  <idno type="IDREF">https://www.idref.fr/220420521</idno>
                  <idno type="VIAF">https://viaf.org/viaf/119151302887948660077</idno>
                  <affiliation ref="#struct-1065372"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Christophe</forename>
                    <surname>Servan</surname>
                  </persName>
                  <email type="md5">65a2a8d7160b6e43a6e9caa24d9af1f2</email>
                  <email type="domain">gmail.com</email>
                  <idno type="idhal" notation="string">servan</idno>
                  <idno type="idhal" notation="numeric">4128</idno>
                  <idno type="halauthorid" notation="string">9504-4128</idno>
                  <idno type="IDREF">https://www.idref.fr/147053609</idno>
                  <idno type="ORCID">https://orcid.org/0000-0003-2306-7075</idno>
                  <affiliation ref="#struct-1065372"/>
                  <affiliation ref="#struct-1222886"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Sophie</forename>
                    <surname>Rosset</surname>
                  </persName>
                  <email type="md5">dbb4013fd09f2b07c31e3a8e3878f7bc</email>
                  <email type="domain">limsi.fr</email>
                  <idno type="idhal" notation="string">sophie-rosset</idno>
                  <idno type="idhal" notation="numeric">14913</idno>
                  <idno type="halauthorid" notation="string">11264-14913</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-6865-4989</idno>
                  <idno type="IDREF">https://www.idref.fr/137157835</idno>
                  <idno type="ISNI">http://isni.org/isni/0000000357034976</idno>
                  <idno type="VIAF">https://viaf.org/viaf/193001246</idno>
                  <affiliation ref="#struct-1061259"/>
                  <affiliation ref="#struct-1065372"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>KG-LLM@LREC  - Knowledge Graphs and Large Language Models</title>
                  <date type="start">2026-05-16</date>
                  <date type="end">2026-05-16</date>
                  <settlement>Palma De Majorque</settlement>
                  <country key="ES">Spain</country>
                </meeting>
                <respStmt>
                  <resp>conferenceOrganizer</resp>
                  <name>ELRA</name>
                </respStmt>
                <imprint/>
              </monogr>
              <idno type="arxiv">2604.02866</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">multilingual NLP</term>
                <term xml:lang="en">propositioner</term>
                <term xml:lang="en">OpenIE</term>
                <term xml:lang="en">triplet extraction</term>
                <term xml:lang="en">knowledge graph</term>
                <term xml:lang="en">atomic propositions</term>
              </keywords>
              <classCode scheme="halDomain" n="info.info-cl">Computer Science [cs]/Computation and Language [cs.CL]</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>&lt;div&gt;&lt;p&gt;Knowledge Graph construction from natural language requires extracting structured triplets from complex, information-dense sentences. In this paper, we investigate if the decomposition of text into atomic propositions (minimal, semantically autonomous units of information) can improve the triplet extraction. We introduce MPropositionneur-V2, a small multilingual model covering six European languages trained by knowledge distillation from Qwen3-32B into a Qwen3-0.6B architecture, and we evaluate its integration into two extraction paradigms: entity-centric (GLiREL) and generative (Qwen3). Experiments on SMiLER, FewRel, DocRED and CaRB show that atomic propositions benefit weaker extractors (GLiREL, CoreNLP, 0.6B models), improving relation recall and, in the multilingual setting, overall accuracy. For stronger LLMs, a fallback combination strategy recovers entity recall losses while preserving the gains in relation extraction. These results show that atomic propositions are an interpretable intermediate data structure that complements extractors without replacing them.&lt;/p&gt;&lt;/div&gt;</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="department" xml:id="struct-1065372" status="VALID">
          <orgName>Sciences et Technologies des Langues - LISN</orgName>
          <orgName type="acronym">STL</orgName>
          <date type="start">2021-01-01</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
          <listRelation>
            <relation active="#struct-1061259" type="direct"/>
            <relation active="#struct-300009" type="indirect"/>
            <relation active="#struct-411575" type="indirect"/>
            <relation active="#struct-419361" type="indirect"/>
            <relation name="UMR9015" active="#struct-441569" type="indirect"/>
          </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="laboratory" xml:id="struct-1061259" status="VALID">
          <idno type="IdRef">253160316</idno>
          <idno type="RNSR">202123711L</idno>
          <idno type="ROR">https://ror.org/00rd81916</idno>
          <idno type="Wikidata">Q109620676</idno>
          <orgName>Laboratoire Interdisciplinaire des Sciences du Numérique</orgName>
          <orgName type="acronym">LISN</orgName>
          <date type="start">2021-01-01</date>
          <desc>
            <address>
              <addrLine>Campus Universitaire bât 507Rue du Belvédère91405 Orsay cedexet Campus Universitaire bât 6401 rue Raimond Castaing91190 Gif-sur-Yvette</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.lisn.upsaclay.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-300009" type="direct"/>
            <relation active="#struct-411575" type="direct"/>
            <relation active="#struct-419361" type="direct"/>
            <relation name="UMR9015" active="#struct-441569" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-1222886" status="VALID">
          <orgName>Agence ministérielle pour l'IA de défense [France]</orgName>
          <orgName type="acronym">AMIAD</orgName>
          <date type="start">2024-05-01</date>
          <desc>
            <address>
              <addrLine>Pôle technique : 136, La Roche Marguerite – 35170 Bruz – France / Pôle recherche : École Polytechnique – 91120 Palaiseau – France</addrLine>
              <country key="FR"/>
            </address>
          </desc>
          <listRelation>
            <relation active="#struct-533790" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-300009" status="VALID">
          <idno type="ROR">https://ror.org/02kvxyf05</idno>
          <orgName>Institut National de Recherche en Informatique et en Automatique</orgName>
          <orgName type="acronym">Inria</orgName>
          <desc>
            <address>
              <addrLine>Domaine de VoluceauRocquencourt - BP 10578153 Le Chesnay Cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.inria.fr/en/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-411575" status="VALID">
          <idno type="IdRef">184443237</idno>
          <idno type="ROR">https://ror.org/019tcpt25</idno>
          <orgName>CentraleSupélec</orgName>
          <desc>
            <address>
              <addrLine>3, rue Joliot Curie,Plateau de Moulon,91192 GIF-SUR-YVETTE Cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.centralesupelec.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="regroupinstitution" xml:id="struct-533790" status="VALID">
          <idno type="IdRef">026382385</idno>
          <idno type="ISNI">0000000121946119</idno>
          <idno type="ROR">https://ror.org/025er3q23</idno>
          <idno type="Wikidata">Q75706</idno>
          <orgName>Ministère des armées – Ministère de la défense [France] (1946-....)</orgName>
          <desc>
            <address>
              <addrLine>60 boulevard du général Martial Valin – 75509 Paris cedex 15</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.defense.gouv.fr/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>