GCRC
收藏资源简介:
GCRC是一个新的挑战性MRC数据集,源自高考中文,用于可解释性评估。该数据集包含超过5000个文本和超过8700个多项选择题(约15000个选项),并标注了句子级支持事实、干扰项的错误原因以及回答问题所需的推理技能。相关实验表明,这个数据集更具挑战性,对于诊断系统的局限性非常有用,并将帮助研究人员开发新的机器学习和推理方法来解决这些未来挑战性问题。
GCRC is a novel and challenging Machine Reading Comprehension (MRC) dataset derived from the Chinese College Entrance Examination, designed for interpretability evaluation. This dataset encompasses over 5,000 texts and more than 8,700 multiple-choice questions (approximately 15,000 options), annotated with sentence-level supporting facts, reasons for distractor errors, and the reasoning skills required to answer the questions. Relevant experiments have demonstrated that this dataset is more challenging, highly useful for diagnosing system limitations, and will assist researchers in developing new machine learning and reasoning methods to address these future challenges.
数据集概述
GCRC(高考中文机器阅读理解数据集)是一个包含超过5000篇文本和超过8700个多项选择题的数据集,旨在通过标准化的人类测试有效且实际地评估机器智能。该数据集特别关注高考阅读理解测试,涵盖了过去10年的试题,并提供了三种类型的标注信息:句子级支持事实、干扰项的错误原因以及回答问题所需的推理技能。
数据集特点
- 数据量:包含超过5000篇文本和8700多个多项选择题。
- 标注信息:提供句子级支持事实、干扰项的错误原因和推理技能的标注。
- 应用目标:旨在通过高考阅读理解任务评估机器智能,并诊断系统的推理过程。
数据集用途
GCRC数据集适用于开发和评估机器阅读理解模型,特别是那些需要解释性评估的模型。该数据集的挑战性有助于研究人员诊断系统限制,并开发新的机器学习和推理方法。
作者和机构
- 作者:Hongye Tan, Xiaoyue Wang, Yu Ji, Ru Li, Xiaoli Li, Zhiwei Hu, Yunxiao Zhao, Xiaoqi Han.
- 机构:Shanxi University
引用信息
bibtex @inproceedings{tan-etal-2021-gcrc, title = "{GCRC}: A New Challenging {MRC} Dataset from {G}aokao {C}hinese for Explainable Evaluation", author = "Tan, Hongye and Wang, Xiaoyue and Ji, Yu and Li, Ru and Li, Xiaoli and Hu, Zhiwei and Zhao, Yunxiao and Han, Xiaoqi", booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021", month = aug, year = "2021", address = "Online", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2021.findings-acl.113", doi = "10.18653/v1/2021.findings-acl.113", pages = "1319--1330", }




