Conference paper Open Access

CMU's Machine Translation System for IWSLT 2019

Srinivasan, Tejas; Sanabria, Ramon; Metze, Florian


DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4.1/metadata.xsd">
  <identifier identifierType="DOI">10.5281/zenodo.3525531</identifier>
  <creators>
    <creator>
      <creatorName>Srinivasan, Tejas</creatorName>
      <givenName>Tejas</givenName>
      <familyName>Srinivasan</familyName>
      <affiliation>Language Technologies Institute, Carnegie Mellon University, USA</affiliation>
    </creator>
    <creator>
      <creatorName>Sanabria, Ramon</creatorName>
      <givenName>Ramon</givenName>
      <familyName>Sanabria</familyName>
      <affiliation>Language Technologies Institute, Carnegie Mellon University, USA</affiliation>
    </creator>
    <creator>
      <creatorName>Metze, Florian</creatorName>
      <givenName>Florian</givenName>
      <familyName>Metze</familyName>
      <affiliation>Language Technologies Institute, Carnegie Mellon University, USA</affiliation>
    </creator>
  </creators>
  <titles>
    <title>CMU's Machine Translation System for IWSLT 2019</title>
  </titles>
  <publisher>Zenodo</publisher>
  <publicationYear>2019</publicationYear>
  <dates>
    <date dateType="Issued">2019-11-02</date>
  </dates>
  <language>en</language>
  <resourceType resourceTypeGeneral="Text">Conference paper</resourceType>
  <alternateIdentifiers>
    <alternateIdentifier alternateIdentifierType="url">https://zenodo.org/record/3525531</alternateIdentifier>
  </alternateIdentifiers>
  <relatedIdentifiers>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsVersionOf">10.5281/zenodo.3525530</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf">https://zenodo.org/communities/iwslt2019</relatedIdentifier>
  </relatedIdentifiers>
  <rightsList>
    <rights rightsURI="https://creativecommons.org/licenses/by/4.0/legalcode">Creative Commons Attribution 4.0 International</rights>
    <rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights>
  </rightsList>
  <descriptions>
    <description descriptionType="Abstract">&lt;p&gt;In Neural Machine Translation (NMT) the usage of sub-􏰃words and characters as source and target units offers a simple and flexible solution for translation of rare and unseen􏰃 words.&amp;nbsp;However, selecting the optimal subword segmentation involves a trade-off between expressiveness and flexibility, and is language and dataset-dependent. We present Block Multitask Learning (BMTL), a novel NMT architecture that predicts multiple targets of different granularities simulta- neously, removing the need to search for the optimal seg- mentation strategy. Our multi-task model exhibits improvements of up to 1.7 BLEU points on each decoder over single-task baseline models with the same number of parameters on datasets from two language pairs of IWSLT15 and one from IWSLT19. The multiple hypotheses generated at different granularities can also be combined as a post-processing step to give better translations.&lt;/p&gt;</description>
  </descriptions>
</resource>
104
75
views
downloads
All versions This version
Views 104104
Downloads 7575
Data volume 63.3 MB63.3 MB
Unique views 101101
Unique downloads 7070

Share

Cite as