This work presents the results of the machine translation (MT) task from the Covid-19 MLIA @ Eval initiative, a community effort to improve the generation of MT systems focused on the current Covid-19 crisis. Nine teams took part in this event, which was divided in two rounds and involved seven different language pairs. Two different scenarios were considered: one in which only the provided data was allowed, and a second one in which the use of external resources was allowed. Overall, best approaches were based on multilingual models and transfer learning, with an emphasis on the importance of applying a cleaning process to the training data.
@article{arxiv.2211.07465,
title = {Findings of the Covid-19 MLIA Machine Translation Task},
author = {Francisco Casacuberta and Alexandru Ceausu and Khalid Choukri and Miltos Deligiannis and Miguel Domingo and Mercedes García-Martínez and Manuel Herranz and Guillaume Jacquet and Vassilis Papavassiliou and Stelios Piperidis and Prokopis Prokopidis and Dimitris Roussis and Marwa Hadj Salah},
journal= {arXiv preprint arXiv:2211.07465},
year = {2022}
}