TransQuest at WMT2020: Sentence-Level Direct Assessment
This work addresses the problem of automated translation quality estimation for machine translation systems, representing an incremental improvement with a winning solution in a specific shared task.
The paper tackled the Sentence-Level Direct Assessment task in WMT 2020 by introducing a simple quality estimation framework based on cross-lingual transformers, achieving state-of-the-art results and winning all language pairs in the competition.
This paper presents the team TransQuest's participation in Sentence-Level Direct Assessment shared task in WMT 2020. We introduce a simple QE framework based on cross-lingual transformers, and we use it to implement and evaluate two different neural architectures. The proposed methods achieve state-of-the-art results surpassing the results obtained by OpenKiwi, the baseline used in the shared task. We further fine tune the QE framework by performing ensemble and data augmentation. Our approach is the winning solution in all of the language pairs according to the WMT 2020 official results.