Beyond english-centric multilingual machine translation

A Fan, S Bhosale, H Schwenk, Z Ma, A El-Kishky… - Journal of Machine …, 2021 - jmlr.org
Existing work in translation demonstrated the potential of massively multilingual machine
translation by training a single model able to translate between any pair of languages …

Learning language specific sub-network for multilingual machine translation

Z Lin, L Wu, M Wang, L Li - ar** and fusing transformer layers for neural machine translation
J Yang, Y Yin, L Yang, S Ma, H Huang… - … on Audio, Speech …, 2022 - ieeexplore.ieee.org
Transformer structure, stacked by a sequence of encoder and decoder network layers,
achieves significant development in neural machine translation. However, vanilla …

Adaptive adapters: An efficient way to incorporate BERT into neural machine translation

J Guo, Z Zhang, L Xu, B Chen… - IEEE/ACM Transactions …, 2021 - ieeexplore.ieee.org
Large-scale pre-trained language models (eg, BERT) have attracted great attention in recent
years. It is straightforward to fine-tune them on natural language understanding tasks such …

Parameter differentiation based multilingual neural machine translation

Q Wang, J Zhang - Proceedings of the AAAI Conference on Artificial …, 2022 - ojs.aaai.org
Multilingual neural machine translation (MNMT) aims to translate multiple languages with a
single model and has been proved successful thanks to effective knowledge transfer among …

Robust optimization for multilingual translation with imbalanced data

X Li, H Gong - Advances in Neural Information Processing …, 2021 - proceedings.neurips.cc
Multilingual models are parameter-efficient and especially effective in improving low-
resource languages by leveraging crosslingual transfer. Despite recent advance in massive …