{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/drop-a-reading-comprehension-benchmark","title":"DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs","arxiv_id":"1903.00161","date":"2019-03-01","proceeding":"NAACL 2019 6","authors":["Dheeru Dua","Yizhong Wang","Pradeep Dasigi","Gabriel Stanovsky","Sameer Singh","Matt Gardner"],"abstract":"Reading comprehension has recently seen rapid progress, with systems matching\nhumans on the most popular datasets for the task. However, a large body of work\nhas highlighted the brittleness of these systems, showing that there is much\nwork left to be done. We introduce a new English reading comprehension\nbenchmark, DROP, which requires Discrete Reasoning Over the content of\nParagraphs. In this crowdsourced, adversarially-created, 96k-question\nbenchmark, a system must resolve references in a question, perhaps to multiple\ninput positions, and perform discrete operations over them (such as addition,\ncounting, or sorting). These operations require a much more comprehensive\nunderstanding of the content of paragraphs than what was necessary for prior\ndatasets. We apply state-of-the-art methods from both the reading comprehension\nand semantic parsing literature on this dataset and show that the best systems\nonly achieve 32.7% F1 on our generalized accuracy metric, while expert human\nperformance is 96.0%. We additionally present a new model that combines reading\ncomprehension methods with simple numerical reasoning to achieve 47.0% F1.","url_abs":"http://arxiv.org/abs/1903.00161v2","url_pdf":"http://arxiv.org/pdf/1903.00161v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"drop-a-reading-comprehension-benchmark","repo_url":"https://github.com/francescomontagna/NAQANet-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"drop-a-reading-comprehension-benchmark","repo_url":"https://github.com/m3yrin/naqanet_notebook","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"drop-a-reading-comprehension-benchmark","repo_url":"https://github.com/allenai/allennlp-reading-comprehension","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"semantic-parsing","task_name":"Semantic Parsing"}],"methods":[],"datasets_introduced":[{"slug":"drop","name":"DROP","full_name":"Discrete Reasoning Over Paragraphs"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-drop-test","task":"Question Answering","dataset":"DROP Test","model":"NAQA Net","rank_in_archive_order":14,"of":16,"metrics":{"F1":"47.01"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-drop-test","task":"Question Answering","dataset":"DROP Test","model":"BERT","rank_in_archive_order":16,"of":16,"metrics":{"F1":"32.7"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.00161","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}