This report describes our VolcTrans system for the WMT22 shared task on large-scale multilingual machine translation. We participated in the unconstrained track which allows the use of external resources. Our system is a transformerbased multilingual model trained on data from multiple sources including the public training set from the data track, NLLB data provided by Meta AI, self-collected parallel corpora, and pseudo bitext from back-translation. A series of heuristic rules clean both bilingual and monolingual texts. On the official test set, our system achieves 17.3 BLEU, 21.9 spBLEU, and 41.9 chrF2++ on average over all language pairs. The average inference speed is 11.5 sentences per second using a single Nvidia Tesla V100 GPU. Our code and trained models are available at https://github.com/xian8/wmt22
@article{arxiv.2210.11599,
title = {The VolcTrans System for WMT22 Multilingual Machine Translation Task},
author = {Xian Qian and Kai Hu and Jiaqiang Wang and Yifeng Liu and Xingyuan Pan and Jun Cao and Mingxuan Wang},
journal= {arXiv preprint arXiv:2210.11599},
year = {2022}
}