{"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/sense-embedding-learning-for-word-sense","title":"Sense Embedding Learning for Word Sense Induction","arxiv_id":"1606.05409","date":"2016-06-17","proceeding":"SEMEVAL 2016 8","authors":["Linfeng Song","Zhiguo Wang","Haitao Mi","Daniel Gildea"],"abstract":"Conventional word sense induction (WSI) methods usually represent each\ninstance with discrete linguistic features or cooccurrence features, and train\na model for each polysemous word individually. In this work, we propose to\nlearn sense embeddings for the WSI task. In the training stage, our method\ninduces several sense centroids (embedding) for each polysemous word. In the\ntesting stage, our method represents each instance as a contextual vector, and\ninduces its sense by finding the nearest sense centroid in the embedding space.\nThe advantages of our method are (1) distributed sense vectors are taken as the\nknowledge representations which are trained discriminatively, and usually have\nbetter performance than traditional count-based distributional models, and (2)\na general model for the whole vocabulary is jointly trained to induce sense\ncentroids under the mutlitask learning framework. Evaluated on SemEval-2010 WSI\ndataset, our method outperforms all participants and most of the recent\nstate-of-the-art methods. We further verify the two advantages by comparing\nwith carefully designed baselines.","url_abs":"http://arxiv.org/abs/1606.05409v2","url_pdf":"http://arxiv.org/pdf/1606.05409v2.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":"word-sense-induction","task_name":"Word Sense Induction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/word-sense-induction-on-semeval-2010-wsi-1","task":"Word Sense Induction","dataset":"SemEval 2010 WSI","model":"SE-WSI-fix","rank_in_archive_order":4,"of":5,"metrics":{"AVG":"23.24","F-Score":"55.1","V-Measure":"9.8"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}