{"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/learning-semantic-representations-for-novel","title":"Learning Semantic Representations for Novel Words: Leveraging Both Form and Context","arxiv_id":"1811.03866","date":"2018-11-09","proceeding":null,"authors":["Timo Schick","Hinrich Schütze"],"abstract":"Word embeddings are a key component of high-performing natural language\nprocessing (NLP) systems, but it remains a challenge to learn good\nrepresentations for novel words on the fly, i.e., for words that did not occur\nin the training data. The general problem setting is that word embeddings are\ninduced on an unlabeled training corpus and then a model is trained that embeds\nnovel words into this induced embedding space. Currently, two approaches for\nlearning embeddings of novel words exist: (i) learning an embedding from the\nnovel word's surface-form (e.g., subword n-grams) and (ii) learning an\nembedding from the context in which it occurs. In this paper, we propose an\narchitecture that leverages both sources of information - surface-form and\ncontext - and show that it results in large increases in embedding quality. Our\narchitecture obtains state-of-the-art results on the Definitional Nonce and\nContextual Rare Words datasets. As input, we only require an embedding set and\nan unlabeled corpus for training our architecture to produce embeddings\nappropriate for the induced embedding space. Thus, our model can easily be\nintegrated into any existing NLP system and enhance its capability to handle\nnovel words.","url_abs":"http://arxiv.org/abs/1811.03866v1","url_pdf":"http://arxiv.org/pdf/1811.03866v1.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":"learning-semantic-representations-for-novel","repo_url":"https://github.com/timoschick/form-context-model","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"form","task_name":"Form"},{"task_slug":"learning-semantic-representations","task_name":"Learning Semantic Representations"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1811.03866","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}