The Ubiqus English-Inuktitut System for WMT20
This paper addresses the challenge of machine translation for low-resource agglutinative languages like Inuktitut, which is a significant problem for language preservation and communication for Inuktitut speakers.
The Ubiqus system for WMT20 English-Inuktitut translation used a multilingual Transformer model trained on several agglutinative languages. This approach aimed to address the challenges of translating Inuktitut, a low-resource agglutinative language.
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.