arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2204.13352cs.CL
收藏

RoBLEURT提交WMT2021审查任务

RoBLEURT Submission for the WMT2021 Metrics Task

University of Macau(澳门大学);DAMO Academy, Alibaba Group(阿里巴巴集团达摩院)

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

摘要:

在本文中,我们提出了我们对共享任务的提交:RoBLEURT(鲁棒优化BLEURT的训练)。在研究了可训练度量的最新进展之后,我们总结了几个至关重要的方面,以获得性能良好的度量模型:1)联合利用包含源模型和仅参考模型的优势,2)使用大量合成数据对持续预训练模型,3)使用数据去噪策略微调模型。实验结果表明,我们的模型在10个英语语言对中的8个上与WMT 2020人类注释达到了最先进的相关性。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑