The paper proposes a new method for training neural machine translation (NMT) models. The method is based on the idea of self-supervised learning, where the model is trained on a large corpus of unlabeled data. The model is then fine-tuned on a small corpus of labeled data. The authors show that their method results in improved translation performance on a variety of benchmarks.
What was the cause of the Hundred Years' War?
Which battle ended the Hundred Years' War?
Which battle was the largest of the Hundred Years' War?
Previous