# UdS Submission for the WMT 19 Automatic Post-Editing Task

## TL;DR

This paper presents a transformer-based approach for automatic post-editing of English-German translations, incorporating context-aware modeling and joint training with a de-noising encoder to improve translation quality.

## Contribution

It introduces an adapted transformer architecture with joint training for APE, leveraging context information and de-noising techniques, which is novel for this task.

## Key findings

- Improved post-editing accuracy over baseline models
- Effective integration of context information in APE
- Demonstrated benefits of joint training with de-noising encoder

## Abstract

In this paper, we describe our submission to the English-German APE shared task at WMT 2019. We utilize and adapt an NMT architecture originally developed for exploiting context information to APE, implement this in our own transformer model and explore joint training of the APE task with a de-noising encoder.