{"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/a-parallel-hierarchical-model-for-machine","title":"A Parallel-Hierarchical Model for Machine Comprehension on Sparse Data","arxiv_id":"1603.08884","date":"2016-03-29","proceeding":"ACL 2016 8","authors":["Adam Trischler","Zheng Ye","Xingdi Yuan","Jing He","Phillip Bachman","Kaheer Suleman"],"abstract":"Understanding unstructured text is a major goal within natural language\nprocessing. Comprehension tests pose questions based on short text passages to\nevaluate such understanding. In this work, we investigate machine comprehension\non the challenging {\\it MCTest} benchmark. Partly because of its limited size,\nprior work on {\\it MCTest} has focused mainly on engineering better features.\nWe tackle the dataset with a neural approach, harnessing simple neural networks\narranged in a parallel hierarchy. The parallel hierarchy enables our model to\ncompare the passage, question, and answer from a variety of trainable\nperspectives, as opposed to using a manually designed, rigid feature set.\nPerspectives range from the word level to sentence fragments to sequences of\nsentences; the networks operate only on word-embedding representations of text.\nWhen trained with a methodology designed to help cope with limited training\ndata, our Parallel-Hierarchical model sets a new state of the art for {\\it\nMCTest}, outperforming previous feature-engineered approaches slightly and\nprevious neural approaches by a significant margin (over 15\\% absolute).","url_abs":"http://arxiv.org/abs/1603.08884v1","url_pdf":"http://arxiv.org/pdf/1603.08884v1.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":"a-parallel-hierarchical-model-for-machine","repo_url":"https://github.com/Maluuba/mctest-model","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-mctest-160","task":"Question Answering","dataset":"MCTest-160","model":"syntax, frame, coreference, and word embedding features","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"75.27%"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-mctest-500","task":"Question Answering","dataset":"MCTest-500","model":"Parallel-Hierarchical","rank_in_archive_order":1,"of":2,"metrics":{"Accuracy":"71%"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-mctest-500","task":"Question Answering","dataset":"MCTest-500","model":"syntax, frame, coreference, and word embedding features","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"69.94%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.08884","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}