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, 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 automatically-generated cloze datasets. We test the performance of dedicatedly designed baseline models including a language model trained on the One Billion Word Corpus and show humans outperform them by a significant margin. We investigate the source of the performance gap, trace model deficiencies to some distinct properties of CLOTH, and identify the limited ability of comprehending the long-term context to be the key bottleneck.
@article{arxiv.1711.03225,
title = {Large-scale Cloze Test Dataset Created by Teachers},
author = {Qizhe Xie and Guokun Lai and Zihang Dai and Eduard Hovy},
journal= {arXiv preprint arXiv:1711.03225},
year = {2018}
}