XL-BEL
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/cambridgeltl/sapbert
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一项跨语言的生物医学实体链接任务，它要求模型能够将任何语言中的提及与UMLS中的语言无关的概念唯一标识符（CUI）相关联。该数据集挑战了在表示领域实体和关联不同语言中实体名称方面的能力，其规模覆盖了10种语言，每种语言包含1000个示例。这项任务的目标是进行跨语言的生物医学实体链接。
This dataset is designed for a cross-lingual biomedical entity linking task, which requires models to associate mentions from any language with the language-neutral concept unique identifiers (CUIs) in the Unified Medical Language System (UMLS). This dataset challenges models' capabilities in domain entity representation and cross-language entity name alignment, covering 10 languages with 1000 examples for each language. The objective of this task is cross-lingual biomedical entity linking.
提供机构：
Derived from Wikipedia and UMLS
