A Framework for Hierarchical Multilingual Machine Translation
This work addresses the challenge of enhancing machine translation for low-resource languages, which is an incremental improvement over existing multilingual strategies.
The paper tackles the problem of improving multilingual machine translation for low-resource languages by proposing a hierarchical framework that uses a typological language family tree to enable transfer among similar languages while avoiding negative effects from dissimilar ones. The result is demonstrated through exhaustive experimentation on a dataset with 41 languages, showing improved performance for low-resource languages.
Multilingual machine translation has recently been in vogue given its potential for improving machine translation performance for low-resource languages via transfer learning. Empirical examinations demonstrating the success of existing multilingual machine translation strategies, however, are limited to experiments in specific language groups. In this paper, we present a hierarchical framework for building multilingual machine translation strategies that takes advantage of a typological language family tree for enabling transfer among similar languages while avoiding the negative effects that result from incorporating languages that are too different to each other. Exhaustive experimentation on a dataset with 41 languages demonstrates the validity of the proposed framework, especially when it comes to improving the performance of low-resource languages via the use of typologically related families for which richer sets of resources are available.