Conference paper Open Access

Generic and Specialized Word Embeddings for Multi-Domain Machine Translation

Pham, MinhQuang; Crego, Josep; Yvon, François; Senellart, Jean


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <controlfield tag="005">20200120165657.0</controlfield>
  <controlfield tag="001">3524979</controlfield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">SYSTRAN / 5 rue Feydeau, 75002 Paris, France</subfield>
    <subfield code="a">Crego, Josep</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">LIMSI, CNRS, Université Paris-Saclay 91405 Orsay, France</subfield>
    <subfield code="a">Yvon, François</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">SYSTRAN / 5 rue Feydeau, 75002 Paris, France</subfield>
    <subfield code="a">Senellart, Jean</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">336061</subfield>
    <subfield code="z">md5:33644a5b7a68b952b82c4e9c6deddc3c</subfield>
    <subfield code="u">https://zenodo.org/record/3524979/files/IWSLT2019_paper_10.pdf</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2019-11-02</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire</subfield>
    <subfield code="p">user-iwslt2019</subfield>
    <subfield code="o">oai:zenodo.org:3524979</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">SYSTRAN / 5 rue Feydeau, 75002 Paris, France &amp; LIMSI, CNRS, Université Paris-Saclay 91405 Orsay, France</subfield>
    <subfield code="a">Pham, MinhQuang</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Generic and Specialized Word Embeddings for Multi-Domain Machine Translation</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-iwslt2019</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">https://creativecommons.org/licenses/by/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution 4.0 International</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;Supervised machine translation works well when the train and test data are sampled from the same distribution. When this is not the case,&amp;nbsp;adaptation&amp;nbsp;techniques help ensure that the knowledge learned from out-of-domain texts generalises to in-domain sentences. We study here a related setting,&amp;nbsp;multi-domain adaptation, where the number of domains is potentially large and adapting separately to each domain would waste training resources. Our proposal transposes to neural machine translation the feature expansion technique of (Daum&amp;eacute;&amp;nbsp;III, 2007): it isolates domain-agnostic from domain-specific lexical representations, while sharing the most of the network across domains. Our experiments use two architectures and two language pairs: they show that our approach, while simple and computationally inexpensive, outperforms several strong baselines and delivers a multi-domain system that successfully translates texts from diverse sources.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="773" ind1=" " ind2=" ">
    <subfield code="n">doi</subfield>
    <subfield code="i">isVersionOf</subfield>
    <subfield code="a">10.5281/zenodo.3524978</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.5281/zenodo.3524979</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">conferencepaper</subfield>
  </datafield>
</record>
92
68
views
downloads
All versions This version
Views 9292
Downloads 6868
Data volume 22.9 MB22.9 MB
Unique views 8181
Unique downloads 6262

Share

Cite as