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

@article{Tandon2019WIQAAD,
  title={WIQA: A dataset for "What if..." reasoning over procedural text},
  author={Niket Tandon and Bhavana Dalvi Mishra and Keisuke Sakaguchi and Antoine Bosselut and Peter Clark},
  journal={ArXiv},
  year={2019},
  volume={abs/1909.04739}
}
We introduce WIQA, the first large-scale dataset of "What if..." questions over procedural text. WIQA contains three parts: a collection of paragraphs each describing a process, e.g., beach erosion; a set of crowdsourced influence graphs for each paragraph, describing how one change affects another; and a large (40k) collection of "What if...?" multiple-choice questions derived from the graphs. For example, given a paragraph about beach erosion, would stormy weather result in more or less… CONTINUE READING

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • We find that state-of-the-art models achieve 73.8% accuracy, well below the human performance of 96.3%.

References

Publications referenced by this paper.