SCDE: Sentence Cloze Dataset with High Quality Distractors From Examinations

Xiang Kong, Varun Gangal, Eduard Hovy


Abstract
We introduce SCDE, a dataset to evaluate the performance of computational models through sentence prediction. SCDE is a human created sentence cloze dataset, collected from public school English examinations. Our task requires a model to fill up multiple blanks in a passage from a shared candidate set with distractors designed by English teachers. Experimental results demonstrate that this task requires the use of non-local, discourse-level context beyond the immediate sentence neighborhood. The blanks require joint solving and significantly impair each other’s context. Furthermore, through ablations, we show that the distractors are of high quality and make the task more challenging. Our experiments show that there is a significant performance gap between advanced models (72%) and humans (87%), encouraging future models to bridge this gap.
Anthology ID:
2020.acl-main.502
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5668–5683
Language:
URL:
https://aclanthology.org/2020.acl-main.502
DOI:
10.18653/v1/2020.acl-main.502
Bibkey:
Cite (ACL):
Xiang Kong, Varun Gangal, and Eduard Hovy. 2020. SCDE: Sentence Cloze Dataset with High Quality Distractors From Examinations. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5668–5683, Online. Association for Computational Linguistics.
Cite (Informal):
SCDE: Sentence Cloze Dataset with High Quality Distractors From Examinations (Kong et al., ACL 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2020.acl-main.502.pdf
Video:
 http://slideslive.com/38928704
Code
 shawnkx/SCDE
Data
SCDECBTCLOTHLAMBADARACEROCStories