nmT5 - Is parallel data still relevant for pre-training massively multilingual language models?

Linting Xue
Mihir Sanjay Kale
Rami Al-Rfou
Annual Meeting of the Association for Computational Linguistics (ACL) (2021) (to appear)
Google Scholar

Abstract

Recently, mT5 - a massively multilingual version of T5 - leveraged a unified text-to-text format to attain state-of-the-art results on a wide variety of multilingual NLP tasks. In this paper, we investigate the impact of incorporating parallel data into mT5 pre-training. We find that simply multi-tasking language modeling with objectives such as machine translation during pre-training leads to improved performance on downstream multilingual and cross-lingual tasks. However, the gains start to diminish as the model capacity increases, suggesting that parallel data might not be as essential for larger models. At the same time, even at larger model sizes, we find that pre-training with parallel data still provides benefits in the limited labelled data regime.