nmT5: Is Parallel Data Still Relevant for Pre-training Massively Multilingual Language Models?
Abstract
nmT5 studies the role of parallel data in pre-training and adaptation of massively multilingual language models.
Type
Publication
Proceedings of the Annual Meeting of the Association for Computational Linguistics
Rami Al-Rfou
Member of Technical Staff - TLM
My research interests include language modeling, embodied AI, motion forecasting, and multilingual modeling.