Skip to content
conftrace_
2021 EMNLP EMNLP 2021

EdinSaar@WMT21: North-Germanic Low-Resource Multilingual NMT

Abstract

AbstractWe describe the EdinSaar submission to the shared task of Multilingual Low-Resource Translation for North Germanic Languages at the Sixth Conference on Machine Translation (WMT2021). We submit multilingual translation models for translations to/from Icelandic (is), Norwegian-Bokmal (nb), and Swedish (sv). We employ various experimental approaches, including multilingual pre-training, back-translation, fine-tuning, and ensembling. In most translation directions, our models outperform other submitted systems.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Natural Language Processing
🐣 Hot Topic Early Bird - low-resource language

Authors

Svetlana Tchistiakova, Jesujoba Alabi, Koel Dutta Chowdhury, Sourav Dutta, Dana Ruiter

Topics

Natural Language Processing > Applications > Machine Translation Machine Learning > Learning Types > Transfer Learning Deep Learning > Learning Types > Multimodal Learning Machine Learning > Learning Types > Multi-Lingual Learning

Keywords

multilingual translation neural machine translation multilingual pretraining low-resource language model ensembling multilingual model north germanic language
Download PDF

More from EMNLP 2021

All EMNLP 2021 papers →
Machine Translation of Low-Resource Indo-European Languages
CUNI systems for WMT21: Multilingual Low-Resource Translation for Indo-European Languages Shared Task
Transfer Learning with Shallow Decoders: BSC at WMT2021’s Multilingual Low-Resource Translation for Indo-European Languages Shared Task
TenTrans Multilingual Low-Resource Translation System for WMT21 Indo-European Languages Task
The University of Maryland, College Park Submission to Large-Scale Multilingual Shared Task at WMT 2021
To Optimize, or Not to Optimize, That Is the Question: TelU-KU Models for WMT21 Large-Scale Multilingual Machine Translation