SemEval 2021 Task 7
收藏arXiv2025-09-30 收录
下载链接：
https://semeval.github.io/SemEval2021/
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了一系列笑话和非笑话内容，其结构特别设计为包含前提和笑点。为了确保笑话和非笑话的质量，数据集经过了清洗和平衡处理，并应用了一系列特定的规则。总体而言，该数据集包含了3,052个标注样本，其中笑话和非笑话各占一半，分别为1,526个，旨在用于幽默识别任务。
This dataset comprises a series of jokes and non-joke content, with a structure specially designed to include both setup and punchline. To guarantee the quality of both jokes and non-jokes, the dataset has undergone cleaning and balancing procedures, with a set of specific rules applied. Overall, this dataset contains 3,052 labeled samples, with jokes and non-jokes each accounting for half of the total, namely 1,526 samples respectively, and it is intended for humor recognition tasks.
提供机构：
SemEval 2021
