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 2010.15535cs.CL
收藏

Unbabel参与WMT20共同任务

Unbabel's Participation in the WMT20 Metrics Shared Task

Unbabel AI(Unbabel人工智能)

Ricardo Rei, Craig Stewart, Catarina Farinha, Alon Lavie

摘要:

我们介绍了Unbabel团队对WMT 2020共享任务的贡献。我们打算参与所有语言对的段级、文档级和系统级轨道,以及“QE as a Metric”轨道。因此,我们参考去年的测试集来说明我们的模型在这些轨道上的结果。我们的意见书建立在最近提出的COMET框架:我们训练几个估计模型回归不同的人类产生的质量分数和一个新的排名模型训练的相对排名从直接评估。我们还提出了一种简单的技术,用于将段级预测转换为文档级分数。总的来说,我们的系统在以前的测试集上对所有语言对都取得了很好的结果,并且在许多情况下创造了一个新的最先进的水平。

英文摘要:

We present the contribution of the Unbabel team to the WMT 2020 Shared Task on Metrics. We intend to participate on the segment-level, document-level and system-level tracks on all language pairs, as well as the 'QE as a Metric' track. Accordingly, we illustrate results of our models in these tracks with reference to test sets from the previous year. Our submissions build upon the recently proposed COMET framework: We train several estimator models to regress on different human-generated quality scores and a novel ranking model trained on relative ranks obtained from Direct Assessments. We also propose a simple technique for converting segment-level predictions into a document-level score. Overall, our systems achieve strong results for all language pairs on previous test sets and in many cases set a new state-of-the-art.

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。

↑