Quoref
收藏资源简介:
displayName: Quoref labelTypes: - Text license: - CC BY 4.0 mediaTypes: - Text paperUrl: https://arxiv.org/pdf/1908.05803v2.pdf publishDate: "2019" publishUrl: https://allenai.org/data/quoref publisher: - University of Washington - Allen Institute for Artificial Intelligence tags: - Text taskTypes: - Machine Reading Comprehension - Reading Comprehension - Coreference Resolution --- # 数据集介绍 ## 简介 Quoref 是一个 QA 数据集,用于测试阅读理解系统的共指推理能力。在这个跨度选择基准包含来自维基百科的 4.7K 段落中的 24K 问题,系统必须先解决硬共指,然后才能在段落中选择适当的跨度来回答问题。 ## 引文 ``` "@article{dasigi2019quoref, title={Quoref: A reading comprehension dataset with questions requiring coreferential reasoning}, author={Dasigi, Pradeep and Liu, Nelson F and Marasovi{\'c}, Ana and Smith, Noah A and Gardner, Matt}, journal={arXiv preprint arXiv:1908.05803}, year={2019} }" ``` ## Download dataset :modelscope-code[]{type="git"}
displayName: Quoref labelTypes: - 文本(Text) license: - 知识共享署名4.0(CC BY 4.0) mediaTypes: - 文本(Text) paperUrl: https://arxiv.org/pdf/1908.05803v2.pdf publishDate: "2019" publishUrl: https://allenai.org/data/quoref publisher: - 华盛顿大学(University of Washington) - 艾伦人工智能研究所(Allen Institute for Artificial Intelligence) tags: - 文本(Text) taskTypes: - 机器阅读理解(Machine Reading Comprehension) - 阅读理解(Reading Comprehension) - 共指消解(Coreference Resolution) --- # 数据集介绍 ## 简介 Quoref 是一个问答(QA)数据集,旨在评测阅读理解系统的共指推理能力。该跨度选择基准包含源自维基百科的4.7K段文本中的24K个问题,系统需先完成共指消解的难例推理,方能在段落中选取合适的文本跨度以回答对应问题。 ## 引文 "@article{dasigi2019quoref, title={Quoref: A reading comprehension dataset with questions requiring coreferential reasoning}, author={Dasigi, Pradeep and Liu, Nelson F and Marasović, Ana and Smith, Noah A and Gardner, Matt}, journal={arXiv preprint arXiv:1908.05803}, year={2019} }" ## 下载数据集 :modelscope-code[]{type="git"}




