{"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/ultra-fine-entity-typing","title":"Ultra-Fine Entity Typing","arxiv_id":"1807.04905","date":"2018-07-13","proceeding":"ACL 2018 7","authors":["Eunsol Choi","Omer Levy","Yejin Choi","Luke Zettlemoyer"],"abstract":"We introduce a new entity typing task: given a sentence with an entity\nmention, the goal is to predict a set of free-form phrases (e.g. skyscraper,\nsongwriter, or criminal) that describe appropriate types for the target entity.\nThis formulation allows us to use a new type of distant supervision at large\nscale: head words, which indicate the type of the noun phrases they appear in.\nWe show that these ultra-fine types can be crowd-sourced, and introduce new\nevaluation sets that are much more diverse and fine-grained than existing\nbenchmarks. We present a model that can predict open types, and is trained\nusing a multitask objective that pools our new head-word supervision with prior\nsupervision from entity linking. Experimental results demonstrate that our\nmodel is effective in predicting entity types at varying granularity; it\nachieves state of the art performance on an existing fine-grained entity typing\nbenchmark, and sets baselines for our newly-introduced datasets. Our data and\nmodel can be downloaded from: http://nlp.cs.washington.edu/entity_type","url_abs":"http://arxiv.org/abs/1807.04905v1","url_pdf":"http://arxiv.org/pdf/1807.04905v1.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":"ultra-fine-entity-typing","repo_url":"https://github.com/uwnlp/open_type","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"entity-linking","task_name":"Entity Linking"},{"task_slug":"entity-typing","task_name":"Entity Typing"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[{"slug":"open-entity-1","name":"Open Entity","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/entity-typing-on-ontonotes-v5-english","task":"Entity Typing","dataset":"Ontonotes v5 (English)","model":"Choi et al. (2018) w augmentation","rank_in_archive_order":4,"of":4,"metrics":{"F1":"32.0","Precision":"47.1","Recall":"24.2"},"uses_additional_data":false},{"leaderboard":"/sota/entity-typing-on-open-entity-1","task":"Entity Typing","dataset":"Open Entity","model":"UFET-biLSTM","rank_in_archive_order":13,"of":13,"metrics":{"F1":"31.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.04905","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}