Netmarble AI Center's WMT21 Automatic Post-Editing Shared Task Submission

TL;DR

This paper presents a novel multi-stage training and data augmentation approach for automatic post-editing of English-German translations, significantly improving translation quality in the WMT21 shared task.

Contribution

It introduces a curriculum training strategy combined with multi-task learning and external data augmentation for enhanced APE performance.

Findings

Significant TER and BLEU improvements on development data

Effective use of multi-task learning with dynamic weighting

Successful leveraging of external translations for better results

Abstract

This paper describes Netmarble's submission to WMT21 Automatic Post-Editing (APE) Shared Task for the English-German language pair. First, we propose a Curriculum Training Strategy in training stages. Facebook Fair's WMT19 news translation model was chosen to engage the large and powerful pre-trained neural networks. Then, we post-train the translation model with different levels of data at each training stages. As the training stages go on, we make the system learn to solve multiple tasks by adding extra information at different training stages gradually. We also show a way to utilize the additional data in large volume for APE tasks. For further improvement, we apply Multi-Task Learning Strategy with the Dynamic Weight Average during the fine-tuning stage. To fine-tune the APE corpus with limited data, we add some related subtasks to learn a unified representation. Finally, for better performance, we leverage external translations as augmented machine translation (MT) during the post-training and fine-tuning. As experimental results show, our APE system significantly improves the translations of provided MT results by -2.848 and +3.74 on the development dataset in terms of TER and BLEU, respectively. It also demonstrates its effectiveness on the test dataset with higher quality than the development dataset.