论文标题
WMT2021指标任务的Robleurt提交
RoBLEURT Submission for the WMT2021 Metrics Task
论文作者
论文摘要
在本文中，我们介绍了共享指标任务的提交：Robleter（强大地优化Bleurt的培训）。在调查了可训练指标的最新进展之后，我们结论了至关重要的几个方面，可以通过以下方式获得表现良好的指标模型：1）共同利用源包含的模型和仅参考模型的优势，2）通过大规模的合成数据配对，并将模型与数据调整为模型策略进行微量的合成数据，并不断预先培训该模型。实验结果表明，我们的模型与WMT2020人类注释达到最新的相关性，在10对英语对中的8分中的8个。
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.
