Skip to content
conftrace_
2021 NAACL NAACL 2021

NRC-CNRC Machine Translation Systems for the 2021 AmericasNLP Shared Task

Abstract

AbstractWe describe the NRC-CNRC systems submitted to the AmericasNLP shared task on machine translation. We submitted systems translating from Spanish into Wixárika, Nahuatl, Rarámuri, and Guaraní. Our best neural machine translation systems used multilingual pretraining, ensembling, finetuning, training on parts of the development data, and subword regularization. We also submitted translation memory systems as a strong baseline.

Authors

Rebecca Knowles, Darlene Stewart, Samuel Larkin, Patrick Littell

Topics

Natural Language Processing > Applications > Machine Translation Natural Language Processing > Resources & Methods > Multilingual NLP

Keywords

neural machine translation multilingual pretraining translation memory subword regularization
Download PDF

More from NAACL 2021

All NAACL 2021 papers →
Peru is Multilingual, Its Machine Translation Should Be Too?
Findings of the AmericasNLP 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas
Open Machine Translation for Low Resource South American Languages (AmericasNLP 2021 Shared Task Contribution)
Low-Resource Machine Translation Using Cross-Lingual Language Model Pretraining
The REPU CS’ Spanish–Quechua Submission to the AmericasNLP 2021 Shared Task on Open Machine Translation
Moses and the Character-Based Random Babbling Baseline: CoAStaL at AmericasNLP 2021 Shared Task