TY - JOUR AU - AB - Idiom 亡羊补牢 Literal To mend the fence after sheep are lost Cloze-style reading comprehension in Chinese is still limited due to the lack of various cor- Meaning Never be late to try pora. In this paper we propose a large-scale Chinese cloze test dataset ChID, which stud- Table 1: An example of metaphor in idiom. The sense ies the comprehension of idiom, a unique lan- of “亡羊补牢” should be inferred figuratively but not guage phenomenon in Chinese. In this corpus, represented literally using the meanings of the four the idioms in a passage are replaced by blank constituent characters. symbols and the correct answer needs to be chosen from well-designed candidate idioms. Idioms 侃侃而谈 口若悬河 We carefully study how the design of candi- date idioms and the representation of idioms To speak with fervour Fluently and Meanings affect the performance of state-of-the-art mod- and assurance eloquently els. Results show that the machine accuracy is Common Speak much and long substantially worse than that of human, indi- cating a large space for further research. Describe the deme- Describe the Difference anor of the speaker eloquence 1 Introduction Table 2: An example of near-synonyms in idiom, where Machine reading comprehension aims TI - ChID: A Large-scale Chinese IDiom Dataset for Cloze Test JF - Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics DO - 10.18653/v1/p19-1075 DA - 2019-01-01 UR - https://www.deepdyve.com/lp/unpaywall/chid-a-large-scale-chinese-idiom-dataset-for-cloze-test-sXAfM0xZzc DP - DeepDyve ER -