arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2307.03115cs.CL
收藏

KoRC:面向深度文本理解的知识导向阅读理解基准

KoRC: Knowledge oriented Reading Comprehension Benchmark for Deep Text Understanding

BNRist(北京信息科学与技术国家研究中心);Tsinghua University(清华大学);University of Chinese Academy of Sciences(中国科学院大学)

Zijun Yao, Yantao Liu, Xin Lv, Shulin Cao, Jifan Yu, Lei Hou, Juanzi Li

AI总结:

本文提出知识导向阅读理解基准KoRC,利用大规模知识库构建知识性问题并以知识库标签为答案,实验显示最强模型F1仅为68.3%和30.0%,表明深度文本理解仍具挑战。

摘要:

深度文本理解要求建立给定文档与其文本之外先验知识之间的联系,近年来许多基准对此给予了强调。然而,这些基准遇到了两个主要局限。一方面,它们大多需要人工标注知识,导致知识覆盖范围有限。另一方面,它们通常使用文本中的选项或片段作为答案,导致答案空间狭窄。为克服这些局限,本文构建了一个新的具有挑战性的基准,名为KoRC。与以往基准相比,KoRC具有两个优势,即广泛的知识覆盖和灵活的答案格式。具体而言,我们利用大规模知识库来指导标注者或大型语言模型(LLMs)构建知识性问题。此外,我们使用知识库中的标签而非文本片段或选项作为最终答案。我们在KoRC上测试了最先进的模型,实验结果表明,最强的基线在分布内和分布外测试集上的F1值分别仅为68.3%和30.0%。这些结果表明,深度文本理解仍然是一个尚未解决的挑战。基准数据集、排行榜和基线方法已在https://github.com/THU-KEG/KoRC发布。

英文摘要:

Deep text understanding, which requires the connections between a given document and prior knowledge beyond its text, has been highlighted by many benchmarks in recent years. However, these benchmarks have encountered two major limitations. On the one hand, most of them require human annotation of knowledge, which leads to limited knowledge coverage. On the other hand, they usually use choices or spans in the texts as the answers, which results in narrow answer space. To overcome these limitations, we build a new challenging benchmark named KoRc in this paper. Compared with previous benchmarks, KoRC has two advantages, i.e., broad knowledge coverage and flexible answer format. Specifically, we utilize massive knowledge bases to guide annotators or large language models (LLMs) to construct knowledgable questions. Moreover, we use labels in knowledge bases rather than spans or choices as the final answers. We test state-of-the-art models on KoRC and the experimental results show that the strongest baseline only achieves 68.3% and 30.0% F1 measure in the in-distribution and out-of-distribution test set, respectively. These results indicate that deep text understanding is still an unsolved challenge. The benchmark dataset, leaderboard, and baseline methods are released in https://github.com/THU-KEG/KoRC.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑