论文标题
彗星：MT评估的神经框架
COMET: A Neural Framework for MT Evaluation
论文作者
论文摘要
我们提出了彗星，这是一种用于培训多语言机器翻译评估模型的神经框架，该模型获得了与人类判断的最新相关性的新水平。我们的框架利用了跨语性审计的语言建模的最新突破，从而导致高度多语言和适应性的MT评估模型，这些模型从源输入和目标语言参考转换中利用信息，以更准确地预测MT质量。为了展示我们的框架，我们培训了三个具有不同类型的人类判断的模型：直接评估，人介导的翻译编辑率和多维质量指标。我们的模型在WMT 2019指标共享任务上实现了新的最新性能，并证明了对高性能系统的鲁棒性。
We present COMET, a neural framework for training multilingual machine translation evaluation models which obtains new state-of-the-art levels of correlation with human judgements. Our framework leverages recent breakthroughs in cross-lingual pretrained language modeling resulting in highly multilingual and adaptable MT evaluation models that exploit information from both the source input and a target-language reference translation in order to more accurately predict MT quality. To showcase our framework, we train three models with different types of human judgements: Direct Assessments, Human-mediated Translation Edit Rate and Multidimensional Quality Metrics. Our models achieve new state-of-the-art performance on the WMT 2019 Metrics shared task and demonstrate robustness to high-performing systems.
