Two-Phase Cross-Lingual Language Model Fine-Tuning for Machine Translation Quality Estimation

Dongjun Lee


Abstract
In this paper, we describe the Bering Lab’s submission to the WMT 2020 Shared Task on Quality Estimation (QE). For word-level and sentence-level translation quality estimation, we fine-tune XLM-RoBERTa, the state-of-the-art cross-lingual language model, with a few additional parameters. Model training consists of two phases. We first pre-train our model on a huge artificially generated QE dataset, and then we fine-tune the model with a human-labeled dataset. When evaluated on the WMT 2020 English-German QE test set, our systems achieve the best result on the target-side of word-level QE and the second best results on the source-side of word-level QE and sentence-level QE among all submissions.
Anthology ID:
2020.wmt-1.118
Volume:
Proceedings of the Fifth Conference on Machine Translation
Month:
November
Year:
2020
Address:
Online
Venue:
WMT
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1024–1028
Language:
URL:
https://aclanthology.org/2020.wmt-1.118
DOI:
Bibkey:
Cite (ACL):
Dongjun Lee. 2020. Two-Phase Cross-Lingual Language Model Fine-Tuning for Machine Translation Quality Estimation. In Proceedings of the Fifth Conference on Machine Translation, pages 1024–1028, Online. Association for Computational Linguistics.
Cite (Informal):
Two-Phase Cross-Lingual Language Model Fine-Tuning for Machine Translation Quality Estimation (Lee, WMT 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/nodalida-main-page/2020.wmt-1.118.pdf
Video:
 https://slideslive.com/38939546