EUROPA

该数据集是一个面向法律领域的多语言关键词生成数据集，其数据来源于欧洲法院的判决书，涵盖了欧盟全部24种官方语言。为了确保评估时关键词的高质量，该数据集在结构设计上精心排除了元信息中的噪声。规模上，该数据集总共包含284,957个实例（输入/关键词对），平均分布在16种语言中的17,833份判决书上。其任务是关键词生成（Kpg）。

This dataset is a multilingual keyword generation dataset tailored for the legal domain. Its data is sourced from the judgments of the European Court of Justice, covering all 24 official languages of the European Union. To ensure high-quality keywords during evaluation, the dataset was meticulously designed to exclude noise from metadata. In terms of scale, the dataset contains a total of 284,957 instances (input/keyword pairs), evenly distributed across 17,833 judgments across 16 languages. Its task is keyword generation (Kpg).