WIQA: A dataset for "What if..." reasoning over procedural text

@inproceedings{Tandon2019WIQAAD,
  title={WIQA: A dataset for "What if..." reasoning over procedural text},
  author={Niket Tandon and B. D. Mishra and Keisuke Sakaguchi and Antoine Bosselut and P. Clark},
  booktitle={EMNLP/IJCNLP},
  year={2019}
}
We introduce WIQA, the first large-scale dataset of "What if..." questions over procedural text. [...] Key Result We find that state-of-the-art models achieve 73.8% accuracy, well below the human performance of 96.3%. We analyze the challenges, in particular tracking chains of influences, and present the dataset as an open challenge to the community.Expand
14 Citations
Logic-Guided Data Augmentation and Regularization for Consistent Question Answering
  • 14
  • Highly Influenced
  • PDF
What-if I ask you to explain: Explaining the effects of perturbations in procedural text
  • 4
  • PDF
QED: A Framework and Dataset for Explanations in Question Answering
  • 6
  • PDF
CURIE: An Iterative Querying Approach for Reasoning About Situations
  • PDF
GLUCOSE: GeneraLized and COntextualized Story Explanations
  • 11
  • PDF
Thinking Like a Skeptic: Defeasible Inference in Natural Language
  • 6
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 15 REFERENCES
Reasoning about Actions and State Changes by Injecting Commonsense Knowledge
  • 45
  • PDF
Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks
  • 800
  • PDF
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
  • 15,760
  • PDF
Tracking the World State with Recurrent Entity Networks
  • 152
  • PDF
Globally Coherent Text Generation with Neural Checklist Models
  • 139
  • PDF
A Structured Learning Approach to Temporal Relation Extraction
  • 41
  • PDF
A Decomposable Attention Model for Natural Language Inference
  • 833
  • PDF
A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting
  • 6,384
  • PDF
A decision-theoretic generalization of on-line learning and an application to boosting
  • 12,226
  • PDF
...
1
2
...