The Ubiqus English-Inuktitut System for WMT20

This paper describes Ubiqus' submission to the WMT20 English-Inuktitut shared news translation task. Our main system, and only submission, is based on a multilingual approach, jointly training a Transformer model on several agglutinative languages. The English-Inuktitut translation task is chal...

Full description

Bibliographic Details
Main Authors: Hernandez, François, Nguyen, Vincent
Format: Article in Journal/Newspaper
Language:unknown
Published: arXiv 2020
Subjects:
Computation and Language cs.CL
FOS Computer and information sciences
inuktitut
Online Access:https://dx.doi.org/10.48550/arxiv.2011.09249
https://arxiv.org/abs/2011.09249
id ftdatacite:10.48550/arxiv.2011.09249
record_format openpolar
spelling ftdatacite:10.48550/arxiv.2011.09249 2023-05-15T16:55:31+02:00 The Ubiqus English-Inuktitut System for WMT20 Hernandez, François Nguyen, Vincent 2020 https://dx.doi.org/10.48550/arxiv.2011.09249 https://arxiv.org/abs/2011.09249 unknown arXiv Creative Commons Attribution Non Commercial No Derivatives 4.0 International https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode cc-by-nc-nd-4.0 CC-BY-NC-ND Computation and Language cs.CL FOS Computer and information sciences Article CreativeWork article Preprint 2020 ftdatacite https://doi.org/10.48550/arxiv.2011.09249 2022-03-10T15:09:53Z This paper describes Ubiqus' submission to the WMT20 English-Inuktitut shared news translation task. Our main system, and only submission, is based on a multilingual approach, jointly training a Transformer model on several agglutinative languages. The English-Inuktitut translation task is challenging at every step, from data selection, preparation and tokenization to quality evaluation down the line. Difficulties emerge both because of the peculiarities of the Inuktitut language as well as the low-resource context. : System Description paper for WMT 2020 English-Inuktitut News Translation Task Article in Journal/Newspaper inuktitut DataCite Metadata Store (German National Library of Science and Technology)
institution Open Polar
collection DataCite Metadata Store (German National Library of Science and Technology)
op_collection_id ftdatacite
language unknown
topic Computation and Language cs.CL
FOS Computer and information sciences
spellingShingle Computation and Language cs.CL
FOS Computer and information sciences
Hernandez, François
Nguyen, Vincent
The Ubiqus English-Inuktitut System for WMT20
topic_facet Computation and Language cs.CL
FOS Computer and information sciences
description This paper describes Ubiqus' submission to the WMT20 English-Inuktitut shared news translation task. Our main system, and only submission, is based on a multilingual approach, jointly training a Transformer model on several agglutinative languages. The English-Inuktitut translation task is challenging at every step, from data selection, preparation and tokenization to quality evaluation down the line. Difficulties emerge both because of the peculiarities of the Inuktitut language as well as the low-resource context. : System Description paper for WMT 2020 English-Inuktitut News Translation Task
format Article in Journal/Newspaper
author Hernandez, François
Nguyen, Vincent
author_facet Hernandez, François
Nguyen, Vincent
author_sort Hernandez, François
title The Ubiqus English-Inuktitut System for WMT20
title_short The Ubiqus English-Inuktitut System for WMT20
title_full The Ubiqus English-Inuktitut System for WMT20
title_fullStr The Ubiqus English-Inuktitut System for WMT20
title_full_unstemmed The Ubiqus English-Inuktitut System for WMT20
title_sort ubiqus english-inuktitut system for wmt20
publisher arXiv
publishDate 2020
url https://dx.doi.org/10.48550/arxiv.2011.09249
https://arxiv.org/abs/2011.09249
genre inuktitut
genre_facet inuktitut
op_rights Creative Commons Attribution Non Commercial No Derivatives 4.0 International
https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode
cc-by-nc-nd-4.0
op_rightsnorm CC-BY-NC-ND
op_doi https://doi.org/10.48550/arxiv.2011.09249
_version_ 1766046532930371584

Similar Items