sleeping-ai/LAION-Debate
收藏Hugging Face2024-07-04 更新2024-06-29 收录
下载链接:
https://hf-mirror.com/datasets/sleeping-ai/LAION-Debate
下载链接
链接失效反馈官方服务:
资源简介:
LAION-Debate是世界上第一个大型竞争性辩论数据集,涵盖了法律、艺术、金融、生物学和气候等多个领域。数据集的大小在10K到100K之间,主要语言为英语。用户可以使用link2media Python库下载数据集,并按照提供的说明进行格式转换和下载。
The LAION-Debate dataset is a large dataset designed for text-classification, summarization, and sentence-similarity tasks. It includes content from various domains such as legal, art, finance, biology, and climate, and is provided in English. The dataset contains between 10,000 and 100,000 entries. The file also provides specific instructions for handling the dataset, including conversion to text format and citing the dataset and associated libraries. The dataset is described as the Worlds first large Competitive Debate dataset.
提供机构:
sleeping-ai
原始信息汇总
数据集概述
基本信息
- 许可证: Apache 2.0
- 任务类别:
- 文本分类
- 摘要生成
- 句子相似度
- 语言: 英语
- 标签:
- 法律
- 艺术
- 金融
- 生物学
- 气候
- 数据集大小: 10K<n<100K
使用说明
- 将CSV文件转换为文本(.txt)文件,以便link2media库处理和下载(移除表头)。
- 文件名反映索引日期和机构名称(CAM = 剑桥联盟,Ox = 牛津联盟)。
- 如果在工作中使用此数据集,请引用此HF数据集。如果使用link2media库,也请引用该库。
- 每两个季度更新一次链接。
引用
@cite{LAION Debate, author = {tawsif ahmed, LAION}, title = {LAION-Debate: Worlds first large Competitive Debate dataset}, year = {2024}, published = https://laion.ai/notes/laion-debate/, }



