The ADAPT System Description for the IWSLT 2018 Basque to English Translation Task
This work addresses translation for low-resource, morphologically-rich languages, but it is incremental as it applies known back-translation methods to a specific task.
The paper tackled the challenge of neural machine translation for low-resource Basque to English by using synthetic data to enhance translation quality, achieving improvements over models trained only on authentic data.
In this paper we present the ADAPT system built for the Basque to English Low Resource MT Evaluation Campaign. Basque is a low-resourced, morphologically-rich language. This poses a challenge for Neural Machine Translation models which usually achieve better performance when trained with large sets of data. Accordingly, we used synthetic data to improve the translation quality produced by a model built using only authentic data. Our proposal uses back-translated data to: (a) create new sentences, so the system can be trained with more data; and (b) translate sentences that are close to the test set, so the model can be fine-tuned to the document to be translated.