UdS Submission for the WMT 19 Automatic Post-Editing Task

08/09/2019

∙

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.

READ FULL TEXT

UdS Submission for the WMT 19 Automatic Post-Editing Task

MS-UEdin Submission to the WMT2018 APE Shared Task: Dual-Source Transformer for Automatic Post-Editing

The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task

The Transference Architecture for Automatic Post-Editing

LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task

Unbabel's Submission to the WMT2019 APE Shared Task: BERT-based Encoder-Decoder for Automatic Post-Editing

Transformer-based Automatic Post-Editing with a Context-Aware Encoding Approach for Multi-Source Inputs

Levenshtein Training for Word-level Quality Estimation

UdS Submission for the WMT 19 Automatic Post-Editing Task

Related Research

MS-UEdin Submission to the WMT2018 APE Shared Task: Dual-Source Transformer for Automatic Post-Editing

The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task

The Transference Architecture for Automatic Post-Editing

LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task

Unbabel's Submission to the WMT2019 APE Shared Task: BERT-based Encoder-Decoder for Automatic Post-Editing

Transformer-based Automatic Post-Editing with a Context-Aware Encoding Approach for Multi-Source Inputs

Levenshtein Training for Word-level Quality Estimation