DFKI-NMT Submission to the WMT19 News Translation Task
Jingyi Zhang, Josef van Genabith · 2019
This paper describes the DFKI-NMT submission to the WMT19 News translation task.We participated in both English-to-German and German-to-English directions.We trained standard Transformer models and adopted various techniques for effectively training our models, including data selection, backtranslation, in-domain fine-tuning and model ensemble.We show that these training techniques improved the performance of our Transformer models up to 5 BLEU points.We give a detailed analysis of the performance of our system.