{"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/embracing-data-abundance-booktest-dataset-for","title":"Embracing data abundance: BookTest Dataset for Reading Comprehension","arxiv_id":"1610.00956","date":"2016-10-04","proceeding":null,"authors":["Ondrej Bajgar","Rudolf Kadlec","Jan Kleindienst"],"abstract":"There is a practically unlimited amount of natural language data available.\nStill, recent work in text comprehension has focused on datasets which are\nsmall relative to current computing possibilities. This article is making a\ncase for the community to move to larger data and as a step in that direction\nit is proposing the BookTest, a new dataset similar to the popular Children's\nBook Test (CBT), however more than 60 times larger. We show that training on\nthe new data improves the accuracy of our Attention-Sum Reader model on the\noriginal CBT test data by a much larger margin than many recent attempts to\nimprove the model architecture. On one version of the dataset our ensemble even\nexceeds the human baseline provided by Facebook. We then show in our own human\nstudy that there is still space for further improvement.","url_abs":"http://arxiv.org/abs/1610.00956v1","url_pdf":"http://arxiv.org/pdf/1610.00956v1.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":"embracing-data-abundance-booktest-dataset-for","repo_url":"https://github.com/facebookresearch/ParlAI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"embracing-data-abundance-booktest-dataset-for","repo_url":"https://github.com/joe-prog/https-github.com-facebookresearch-ParlAI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-18","how":"tree_404+repo_404"}}],"tasks":[{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"}],"methods":[],"datasets_introduced":[{"slug":"booktest","name":"BookTest","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1610.00956","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}