{"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/zero-shot-relation-extraction-via-reading","title":"Zero-Shot Relation Extraction via Reading Comprehension","arxiv_id":"1706.04115","date":"2017-06-13","proceeding":"CONLL 2017 8","authors":["Omer Levy","Minjoon Seo","Eunsol Choi","Luke Zettlemoyer"],"abstract":"We show that relation extraction can be reduced to answering simple reading\ncomprehension questions, by associating one or more natural-language questions\nwith each relation slot. This reduction has several advantages: we can (1)\nlearn relation-extraction models by extending recent neural\nreading-comprehension techniques, (2) build very large training sets for those\nmodels by combining relation-specific crowd-sourced questions with distant\nsupervision, and even (3) do zero-shot learning by extracting new relation\ntypes that are only specified at test-time, for which we have no labeled\ntraining examples. Experiments on a Wikipedia slot-filling task demonstrate\nthat the approach can generalize to new questions for known relation types with\nhigh accuracy, and that zero-shot generalization to unseen relation types is\npossible, at lower accuracy levels, setting the bar for future work on this\ntask.","url_abs":"http://arxiv.org/abs/1706.04115v1","url_pdf":"http://arxiv.org/pdf/1706.04115v1.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":"zero-shot-relation-extraction-via-reading","repo_url":"https://github.com/stonybrooknlp/musique","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"zero-shot-relation-extraction-via-reading","repo_url":"https://github.com/zhuzhicai/SQuAD2.0-Baseline-Test-with-BiDAF-No-Answer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"},{"task_slug":"slot-filling","task_name":"Slot Filling"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"},{"task_slug":"zero-shot-generalization","task_name":"Zero-shot Generalization"},{"task_slug":"slot-filling-1","task_name":"slot-filling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.04115","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}