{"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/contextualized-word-representations-for","title":"Contextualized Word Representations for Reading Comprehension","arxiv_id":"1712.03609","date":"2017-12-10","proceeding":"NAACL 2018 6","authors":["Shimi Salant","Jonathan Berant"],"abstract":"Reading a document and extracting an answer to a question about its content\nhas attracted substantial attention recently. While most work has focused on\nthe interaction between the question and the document, in this work we evaluate\nthe importance of context when the question and document are processed\nindependently. We take a standard neural architecture for this task, and show\nthat by providing rich contextualized word representations from a large\npre-trained language model as well as allowing the model to choose between\ncontext-dependent and context-independent word representations, we can obtain\ndramatic improvements and reach performance comparable to state-of-the-art on\nthe competitive SQuAD dataset.","url_abs":"http://arxiv.org/abs/1712.03609v4","url_pdf":"http://arxiv.org/pdf/1712.03609v4.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":"contextualized-word-representations-for","repo_url":"https://github.com/shimisalant/CWR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"RaSoR + TR + LM (single model)","rank_in_archive_order":99,"of":213,"metrics":{"EM":"77.583","F1":"84.163"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"RaSoR + TR (single model)","rank_in_archive_order":118,"of":213,"metrics":{"EM":"75.789","F1":"83.261"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.03609","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}