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.
@article{arxiv.2204.13352,
title = {RoBLEURT Submission for the WMT2021 Metrics Task},
author = {Yu Wan and Dayiheng Liu and Baosong Yang and Tianchi Bi and Haibo Zhang and Boxing Chen and Weihua Luo and Derek F. Wong and Lidia S. Chao},
journal= {arXiv preprint arXiv:2204.13352},
year = {2022}
}