GenericsKB

该数据集是一个庞大的知识库，包含了从多个语料库中收集的超过350万个通用陈述句。这些陈述句自然出现，并标注了主题术语、上下文以及置信度。此外，数据集还包含了“GenericsKB-Best”子集，其中包含超过100万个高质量的通用陈述。规模上，该数据集拥有超过350万个句子，适用于问答和解释生成任务。

This dataset is a large-scale knowledge base containing over 3.5 million generic declarative sentences collected from multiple corpora. These naturally occurring sentences are annotated with topic terms, context, and confidence scores. Additionally, the dataset includes the "GenericsKB-Best" subset, which contains over 1 million high-quality generic statements. With a total of over 3.5 million sentences, this dataset is suitable for question answering and explanation generation tasks.