Preservation of domain knowledge from the source to target is crucial in any translation
workflow. It is common in the translation industry to receive highly specialized projects,
where there is hardly any parallel in-domain data. In such scenarios where there is insufficient
in-domain data to fine-tune Machine Translation (MT) models, producing translations that
are consistent with the relevant context is challenging. In this work, we propose a novel
approach to domain adaptation leveraging state-of-the-art pretrained language models (LMs)
for domain-specific data augmentation for MT, simulating the domain characteristics of
either (a) a small bilingual dataset, or (b) the monolingual source text to be translated.
Combining this idea with back-translation, we can generate huge amounts of synthetic
bilingual in-domain data for both use cases. For our investigation, we use the state-of-the-art
Transformer architecture. We employ mixed fine-tuning to train models that significantly
improve translation of in-domain texts. More specifically, in both scenarios, our proposed
methods achieve improvements of approximately 5-6 BLEU and 2-3 BLEU, respectively, on
the Arabic-to-English and English-to-Arabic language pairs. Furthermore, the outcome of
human evaluation corroborates the automatic evaluation results.
Proceedings of the 15th Biennial Conference of the Association for Machine Translation in the Americas.
1.
Association for Machine Translation in the Americas.
Publisher:
Association for Machine Translation in the Americas
Science Foundation Ireland Centre for Research Training in Digitally-Enhanced Reality (d-real) under Grant No. 18/CRT/6224, Science Foundation Ireland (SFI) Research Centres Programme (Grant No. 13/RC/2106), European Regional Development Fund, and Microsoft Research.
ID Code:
28322
Deposited On:
10 May 2023 13:58 by Thomas Murtagh. Last Modified 10 May 2023 13:58