RoBLEURT Submission for WMT2021 Metrics Task

Yu Wan, Dayiheng Liu, Baosong Yang, Tianchi Bi, Haibo Zhang, Boxing Chen, Weihua Luo, Derek F. Wong, Lidia S. Chao


Abstract
In this paper, we present our submission to Shared Metrics Task: RoBLEURT (Robustly Optimizing the training of BLEURT). After investigating the recent advances of trainable metrics, we conclude several aspects of vital importance to obtain a well-performed metric model by: 1) jointly leveraging the advantages of source-included model and reference-only model, 2) continuously pre-training the model with massive synthetic data pairs, and 3) fine-tuning the model with data denoising strategy. Experimental results show that our model reaching state-of-the-art correlations with the WMT2020 human annotations upon 8 out of 10 to-English language pairs.
Anthology ID:
2021.wmt-1.114
Volume:
Proceedings of the Sixth Conference on Machine Translation
Month:
November
Year:
2021
Address:
Online
Venues:
EMNLP | WMT
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1053–1058
Language:
URL:
https://aclanthology.org/2021.wmt-1.114
DOI:
Bibkey:
Cite (ACL):
Yu Wan, Dayiheng Liu, Baosong Yang, Tianchi Bi, Haibo Zhang, Boxing Chen, Weihua Luo, Derek F. Wong, and Lidia S. Chao. 2021. RoBLEURT Submission for WMT2021 Metrics Task. In Proceedings of the Sixth Conference on Machine Translation, pages 1053–1058, Online. Association for Computational Linguistics.
Cite (Informal):
RoBLEURT Submission for WMT2021 Metrics Task (Wan et al., WMT 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.wmt-1.114.pdf
Data
WMT 2020