Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning
Pradeep Dasigi, Nelson F. Liu, Ana Marasović, Noah A. Smith, Matt Gardner
Abstract
Machine comprehension of texts longer than a single sentence often requires coreference resolution. However, most current reading comprehension benchmarks do not contain complex coreferential phenomena and hence fail to evaluate the ability of models to resolve coreference. We present a new crowdsourced dataset containing more than 24K span-selection questions that require resolving coreference among entities in over 4.7K English paragraphs from Wikipedia. Obtaining questions focused on such phenomena is challenging, because it is hard to avoid lexical cues that shortcut complex reasoning. We deal with this issue by using a strong baseline model as an adversary in the crowdsourcing loop, which helps crowdworkers avoid writing questions with exploitable surface cues. We show that state-of-the-art reading comprehension models perform significantly worse than humans on this benchmark—the best model performance is 70.5 F1, while the estimated human performance is 93.4 F1.- Anthology ID:
- D19-1606
- Volume:
- Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
- Month:
- November
- Year:
- 2019
- Address:
- Hong Kong, China
- Editors:
- Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
- Venues:
- EMNLP | IJCNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5925–5932
- Language:
- URL:
- https://aclanthology.org/D19-1606
- DOI:
- 10.18653/v1/D19-1606
- Cite (ACL):
- Pradeep Dasigi, Nelson F. Liu, Ana Marasović, Noah A. Smith, and Matt Gardner. 2019. Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 5925–5932, Hong Kong, China. Association for Computational Linguistics.
- Cite (Informal):
- Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning (Dasigi et al., EMNLP-IJCNLP 2019)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-1/D19-1606.pdf
- Data
- Quoref, DROP, RACE, SQuAD