David Ifeoluwa Adelani, Jesujoba Oluwadara Alabi, Angela Fan, Julia Kreutzer, Xiaoyu Shen, Machel Reid, Dana Ruiter, D. Klakow, Peter Nabende, Ernie Chang, T. Gwadabe, Freshia Sackey, Bonaventure F. P. Dossou, Chris C. Emezue, Colin Leong, Michael Beukman, Shamsuddeen Hassan Muhammad, Guyo Dub Jarso, Oreen Yousuf, Andre Niyongabo Rubungo, Gilles Hacheme, Eric Peter Wairagala, Muhammad Umair Nasir, Benjamin Ayoade Ajibade, T. Ajayi, Yvonne Wambui Gitau, Jade Z. Abbott, Mohamed Ahmed, Millicent Ochieng, Anuoluwapo Aremu, Perez Ogayo, Jonathan Mukiibi, F. Kabore, Godson Kalipe, Derguene Mbaye, A. Tapo, V. M. Koagne, Edwin Munkoh-Buabeng, Valencia Wagner, Idris Abdulmumin, Ayodele Awokoya, Happy Buzaaba, Blessing K. Sibanda, Andiswa Bukula, Sam Manthalu

NAACL

Paper

Abstract

Recent advances in the pre-training for language models leverage large-scale datasets to create multilingual models. However, low-resource languages are mostly left out in these datasets. This is primarily because many widely spoken languages that are not well represented on the web and therefore excluded from the large-scale crawls for datasets. Furthermore, downstream users of these models are restricted to the selection of languages originally chosen for pre-training. This work investigates how to optimally leverage existing pre-trained models to create low-resource translation systems for 16 African languages. We focus on two questions: 1) How can pre-trained models be used for languages not included in the initial pretraining? and 2) How can the resulting translation models effectively transfer to new domains? To answer these questions, we create a novel African news corpus covering 16 languages, of which eight languages are not part of any existing evaluation dataset. We demonstrate that the most effective strategy for transferring both additional languages and additional domains is to leverage small quantities of high-quality translation data to fine-tune large pre-trained models.