{"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/high-risk-learning-acquiring-new-word-vectors","title":"High-risk learning: acquiring new word vectors from tiny data","arxiv_id":"1707.06556","date":"2017-07-20","proceeding":"EMNLP 2017 9","authors":["Aurelie Herbelot","Marco Baroni"],"abstract":"Distributional semantics models are known to struggle with small data. It is\ngenerally accepted that in order to learn 'a good vector' for a word, a model\nmust have sufficient examples of its usage. This contradicts the fact that\nhumans can guess the meaning of a word from a few occurrences only. In this\npaper, we show that a neural language model such as Word2Vec only necessitates\nminor modifications to its standard architecture to learn new terms from tiny\ndata, using background knowledge from a previously learnt semantic space. We\ntest our model on word definitions and on a nonce task involving 2-6 sentences'\nworth of context, showing a large increase in performance over state-of-the-art\nmodels on the definitional task.","url_abs":"http://arxiv.org/abs/1707.06556v1","url_pdf":"http://arxiv.org/pdf/1707.06556v1.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":"high-risk-learning-acquiring-new-word-vectors","repo_url":"https://github.com/minimalparts/nonce2vec","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"small-data","task_name":"Small Data Image Classification"},{"task_slug":"high","task_name":"Vocal Bursts Intensity Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.06556","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}