conference-paper Open access

IST-Unbabel Participation in the WMT20 Quality Estimation Shared Task

Research footprint

At a glance

Citations
18
References
12
Comments
0
Paper overview

Abstract

We present the joint contribution of IST and Unbabel to the WMT 2020 Shared Task on Quality Estimation.Our team participated on all tracks (Direct Assessment, Post-Editing Effort, Document-Level), encompassing a total of 14 submissions.Our submitted systems were developed by extending the OpenKiwi framework to a transformer-based predictorestimator architecture, and to cope with glassbox, uncertainty-based features coming from neural machine translation systems.

Record transparency

Publication details

DOI
10.18653/v1/2020.wmt-1.119
OpenAlex
W3120791402
Document type
conference-paper
Language
EN
Last metadata update
Community

Comments

Log in to join the discussion.

  1. No comments yet. Start the discussion.