Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning

@inproceedings{Dasigi2019QuorefAR,
  title={Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning},
  author={Pradeep Dasigi and Nelson F. Liu and Ana Marasovi{\'c} and Noah A. Smith and Matt Gardner},
  booktitle={EMNLP/IJCNLP},
  year={2019}
}
Machine comprehension of texts longer than a single sentence often requires coreference resolution. [...] Key Method 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.Expand
Coreference Reasoning in Machine Reading Comprehension
On Making Reading Comprehension More Comprehensive
IIRC: A Dataset of Incomplete Information Reading Comprehension Questions
Why Machine Reading Comprehension Models Learn Shortcuts?
CorefQA: Coreference Resolution as Query-based Span Prediction
Comprehensive Multi-Dataset Evaluation of Reading Comprehension
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 32 REFERENCES
MCTest: A Challenge Dataset for the Open-Domain Machine Comprehension of Text
PreCo: A Large-scale Dataset in Preschool Vocabulary for Coreference Resolution
SQuAD: 100, 000+ Questions for Machine Comprehension of Text
Annotation Artifacts in Natural Language Inference Data
MS MARCO: A Human Generated MAchine Reading COmprehension Dataset
...
1
2
3
4
...