Login (DCU Staff Only)
Login (DCU Staff Only)

DORAS | DCU Research Repository

Explore open access research and scholarly works from DCU

Advanced Search

The ADAPT Centre’s neural MT systems for the WAT 2020 document-level translation task

Jooste, Wandri, Haque, Rejwanul orcid logoORCID: 0000-0003-1680-0099 and Way, Andy orcid logoORCID: 0000-0001-5736-5930 (2020) The ADAPT Centre’s neural MT systems for the WAT 2020 document-level translation task. In: 7th Workshop on Asian Translation (WAT2020), 4 Dec 2020, Suzhou, China (Online).

Abstract
In this paper we describe the ADAPT Centre’s (Team ID: adapt-dcu) submissions to the WAT 2020 document-level Business Scene Dialogue (BSD) translation task. We only considered translating from Japanese to English for this task and secured the third position in the competition as per the rankings of the MT systems based on the human evaluation scores. The machine translation (MT) systems that we built for this task are state-of-the-art Trans- former models. In order to improve the translation quality of our neural MT (NMT) systems, we made use of both in-domain and out-of- domain data for training. We applied various data augmentation techniques for fine-tuning the model parameters. This paper outlines the experiments we carried out for this task and reports the MT systems’ performance on the evaluation test set.
Metadata
Item Type:Conference or Workshop Item (Paper)
Event Type:Workshop
Refereed:Yes
Subjects:Computer Science > Computational linguistics
Computer Science > Machine translating
DCU Faculties and Centres:DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Research Initiatives and Centres > ADAPT
Published in: Proceedings of the 7th Workshop on Asian Translation (WAT2020). . Association for Computational Linguistics (ACL).
Publisher:Association for Computational Linguistics (ACL)
Official URL:https://www.aclweb.org/anthology/2020.wat-1.17
Copyright Information:© 2020 The Authors.
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
ID Code:25205
Deposited On:04 Dec 2020 14:27 by Wandri Jooste . Last Modified 08 Apr 2021 16:32
Documents

Full text available as:

[thumbnail of WAT_2020.pdf]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
113kB
Downloads

Downloads

Downloads per month over past year

Archive Staff Only: edit this record