{"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/breaking-sticks-and-ambiguities-with-adaptive","title":"Breaking Sticks and Ambiguities with Adaptive Skip-gram","arxiv_id":"1502.07257","date":"2015-02-25","proceeding":null,"authors":["Sergey Bartunov","Dmitry Kondrashkin","Anton Osokin","Dmitry Vetrov"],"abstract":"Recently proposed Skip-gram model is a powerful method for learning\nhigh-dimensional word representations that capture rich semantic relationships\nbetween words. However, Skip-gram as well as most prior work on learning word\nrepresentations does not take into account word ambiguity and maintain only\nsingle representation per word. Although a number of Skip-gram modifications\nwere proposed to overcome this limitation and learn multi-prototype word\nrepresentations, they either require a known number of word meanings or learn\nthem using greedy heuristic approaches. In this paper we propose the Adaptive\nSkip-gram model which is a nonparametric Bayesian extension of Skip-gram\ncapable to automatically learn the required number of representations for all\nwords at desired semantic resolution. We derive efficient online variational\nlearning algorithm for the model and empirically demonstrate its efficiency on\nword-sense induction task.","url_abs":"http://arxiv.org/abs/1502.07257v2","url_pdf":"http://arxiv.org/pdf/1502.07257v2.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":"breaking-sticks-and-ambiguities-with-adaptive","repo_url":"https://github.com/sbos/AdaGram.jl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"breaking-sticks-and-ambiguities-with-adaptive","repo_url":"https://github.com/hjian42/Geo-Twitter2019","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"breaking-sticks-and-ambiguities-with-adaptive","repo_url":"https://github.com/lopuhin/python-adagram","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"word-sense-induction","task_name":"Word Sense Induction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1502.07257","atlas_url":"https://app.syntology.ai/?focus=1502.07257","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}