Compositional Questions Do Not Necessitate Multi-hop Reasoning

@inproceedings{Min2019CompositionalQD,
  title={Compositional Questions Do Not Necessitate Multi-hop Reasoning},
  author={Sewon Min and Eric Wallace and Sameer Singh and Matt Gardner and Hannaneh Hajishirzi and Luke Zettlemoyer},
  booktitle={ACL},
  year={2019}
}
Multi-hop reading comprehension (RC) questions are challenging because they require reading and reasoning over multiple paragraphs. We argue that it can be difficult to construct large multi-hop RC datasets. For example, even highly compositional questions can be answered with a single hop if they target specific entity types, or the facts needed to answer them are redundant. Our analysis is centered on HotpotQA, where we show that single-hop reasoning can solve much more of the dataset than… CONTINUE READING

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 15 CITATIONS

Break It Down: A Question Understanding Benchmark

VIEW 6 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Unsupervised Question Decomposition for Question Answering

VIEW 17 EXCERPTS
CITES METHODS, BACKGROUND & RESULTS
HIGHLY INFLUENCED

Answering Complex Open-domain Questions Through Iterative Query Generation

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Do Multi-hop Readers Dream of Reasoning Chains?

VIEW 3 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Hierarchical Graph Network for Multi-hop Question Answering

VIEW 5 EXCERPTS
CITES METHODS, RESULTS & BACKGROUND
HIGHLY INFLUENCED

TRANSFORMER-XH: MULTI-HOP QUESTION AN-

  • SWERING WITH EXTRA, HOP ATTENTION
  • 2019
VIEW 9 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

References

Publications referenced by this paper.
SHOWING 1-10 OF 19 REFERENCES

The Web as a Knowledge-base for Answering Complex Questions

VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Constructing Datasets for Multi-hop Reading Comprehension Across Documents

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

Automatic differentiation in PyTorch

VIEW 1 EXCERPT