Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing 2018
DOI: 10.18653/v1/d18-1257
|View full text |Cite
|
Sign up to set email alerts
|

Large-scale Cloze Test Dataset Created by Teachers

Abstract: Cloze tests are widely adopted in language exams to evaluate students' language proficiency. In this paper, we propose the first large-scale human-created cloze test dataset CLOTH 1 2 , containing questions used in middle-school and high-school language exams. With missing blanks carefully created by teachers and candidate choices purposely designed to be nuanced, CLOTH requires a deeper language understanding and a wider attention span than previously automaticallygenerated cloze datasets. We test the perform… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
46
0
3

Year Published

2018
2018
2023
2023

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 64 publications
(49 citation statements)
references
References 35 publications
(50 reference statements)
0
46
0
3
Order By: Relevance
“…Question Answering over Documents There has been rapid progress in the task of question answering (QA) over documents along with vari- et al, 2015;Trischler et al, 2016), fictional stories (Richardson et al, 2013;Kočiskỳ et al, 2017), and textbooks (Lai et al, 2017;Xie et al, 2017). Many neural QA models have successfully addressed these tasks by leveraging coattention or bidirectional attention mechanisms (Xiong et al, 2018;Seo et al, 2017) to model the codependent context over the document and the question.…”
Section: Related Workmentioning
confidence: 99%
“…Question Answering over Documents There has been rapid progress in the task of question answering (QA) over documents along with vari- et al, 2015;Trischler et al, 2016), fictional stories (Richardson et al, 2013;Kočiskỳ et al, 2017), and textbooks (Lai et al, 2017;Xie et al, 2017). Many neural QA models have successfully addressed these tasks by leveraging coattention or bidirectional attention mechanisms (Xiong et al, 2018;Seo et al, 2017) to model the codependent context over the document and the question.…”
Section: Related Workmentioning
confidence: 99%
“…In most datasets, the answer can be directly found from context. CLOTH (Xie et al, 2018) has a similar setting to ChID, where the answer should be selected from given choices. However, CLOTH is collected from English examinations for secondary/high school students, whose size is limited because documents, blanks, and options are all manually created.…”
Section: Related Workmentioning
confidence: 99%
“…-CLOTH Different from above automatically-generated datasets, CLOTH [93] (CLOze test by TeacHers) is human-created, which is collected from English exams for Chinese students. Questions in the CLOTH are well-designed by middle-school and high-school teachers to examine students' language proficiency including vocabulary, reasoning and grammar.…”
Section: Ms Marco[51]mentioning
confidence: 99%
“…-RACE Like CLOTH dataset [93], RACE [36] is also collected from the English exams for middle school and high school Chinese students. This corpus allows types of passages to be more various.…”
Section: -Mctestmentioning
confidence: 99%