Task Categories: question-answering
Languages: en
Multilinguality: monolingual
Size Categories: 10K<n<100K
Language Creators: crowdsourced
Annotations Creators: crowdsourced
Source Datasets: original

Dataset Card for "drop"

Dataset Summary

DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs. . DROP is a crowdsourced, adversarially-created, 96k-question benchmark, in which a system must resolve references in a question, perhaps to multiple input positions, and perform discrete operations over them (such as addition, counting, or sorting). These operations require a much more comprehensive understanding of the content of paragraphs than what was necessary for prior datasets.

Supported Tasks and Leaderboards

More Information Needed


More Information Needed

Dataset Structure

We show detailed information for up to 5 configurations of the dataset.

Data Instances


  • Size of downloaded dataset files: 7.92 MB
  • Size of the generated dataset: 105.77 MB
  • Total amount of disk used: 113.69 MB

An example of 'validation' looks as follows.

This example was too long and was cropped:

    "answers_spans": {
        "spans": ["Chaz Schilens"]
    "passage": "\" Hoping to rebound from their loss to the Patriots, the Raiders stayed at home for a Week 16 duel with the Houston Texans.  Oak...",
    "question": "Who scored the first touchdown of the game?"

Data Fields

The data fields are the same among all splits.


  • passage: a string feature.
  • question: a string feature.
  • answers_spans: a dictionary feature containing:
    • spans: a string feature.

Data Splits

name train validation
default 77409 9536

Dataset Creation

Curation Rationale

More Information Needed

Source Data

Initial Data Collection and Normalization

More Information Needed

Who are the source language producers?

More Information Needed


Annotation process

More Information Needed

Who are the annotators?

More Information Needed

Personal and Sensitive Information

More Information Needed

Considerations for Using the Data

Social Impact of Dataset

More Information Needed

Discussion of Biases

More Information Needed

Other Known Limitations

More Information Needed

Additional Information

Dataset Curators

More Information Needed

Licensing Information

More Information Needed

Citation Information

  author={Dheeru Dua and Yizhong Wang and Pradeep Dasigi and Gabriel Stanovsky and Sameer Singh and Matt Gardner},
  title={  {DROP}: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs},
  booktitle={Proc. of NAACL},


Thanks to @patrickvonplaten, @thomwolf, @mariamabarham, @lewtun for adding this dataset.

Models trained or fine-tuned on drop