WMT 2020 Human Evaluation Dataset

该数据集包含了来自WMT 2020共享任务的机器翻译输出在段落级别的评分，这些评分是由能够接触到完整文档上下文的职业翻译人员进行评估的。数据集内包含了翻译评分，这些评分以0-100的刻度进行，同时还包括了跨评分者的段落级别平均分数，以及基于多维度质量度量（MQM）框架的错误分析。该数据集的任务是进行机器翻译评估。

This dataset contains paragraph-level scores of machine translation outputs from the WMT 2020 shared task, which were evaluated by professional translators with access to full document context. The dataset includes translation scores rated on a 0-100 scale, as well as inter-annotator paragraph-level average scores and error analysis based on the Multidimensional Quality Metric (MQM) framework. The task addressed by this dataset is machine translation evaluation.