{"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/broad-context-language-modeling-as-reading","title":"Broad Context Language Modeling as Reading Comprehension","arxiv_id":"1610.08431","date":"2016-10-26","proceeding":"EACL 2017 4","authors":["Zewei Chu","Hai Wang","Kevin Gimpel","David Mcallester"],"abstract":"Progress in text understanding has been driven by large datasets that test\nparticular capabilities, like recent datasets for reading comprehension\n(Hermann et al., 2015). We focus here on the LAMBADA dataset (Paperno et al.,\n2016), a word prediction task requiring broader context than the immediate\nsentence. We view LAMBADA as a reading comprehension problem and apply\ncomprehension models based on neural networks. Though these models are\nconstrained to choose a word from the context, they improve the state of the\nart on LAMBADA from 7.3% to 49%. We analyze 100 instances, finding that neural\nnetwork readers perform well in cases that involve selecting a name from the\ncontext based on dialogue or discourse cues but struggle when coreference\nresolution or external knowledge is needed.","url_abs":"http://arxiv.org/abs/1610.08431v3","url_pdf":"http://arxiv.org/pdf/1610.08431v3.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":[],"tasks":[{"task_slug":"coreference-resolution","task_name":"Coreference Resolution"},{"task_slug":"lambada","task_name":"LAMBADA"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"coreference-resolution-1","task_name":"coreference-resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/language-modelling-on-lambada","task":"Language Modelling","dataset":"LAMBADA","model":"Gated-Attention Reader (+ features)","rank_in_archive_order":32,"of":37,"metrics":{"Accuracy":"49.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1610.08431","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}