{"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/lifelong-domain-word-embedding-via-meta","title":"Lifelong Domain Word Embedding via Meta-Learning","arxiv_id":"1805.09991","date":"2018-05-25","proceeding":null,"authors":["Hu Xu","Bing Liu","Lei Shu","Philip S. Yu"],"abstract":"Learning high-quality domain word embeddings is important for achieving good\nperformance in many NLP tasks. General-purpose embeddings trained on\nlarge-scale corpora are often sub-optimal for domain-specific applications.\nHowever, domain-specific tasks often do not have large in-domain corpora for\ntraining high-quality domain embeddings. In this paper, we propose a novel\nlifelong learning setting for domain embedding. That is, when performing the\nnew domain embedding, the system has seen many past domains, and it tries to\nexpand the new in-domain corpus by exploiting the corpora from the past domains\nvia meta-learning. The proposed meta-learner characterizes the similarities of\nthe contexts of the same word in many domain corpora, which helps retrieve\nrelevant data from the past domains to expand the new domain corpus.\nExperimental results show that domain embeddings produced from such a process\nimprove the performance of the downstream tasks.","url_abs":"http://arxiv.org/abs/1805.09991v1","url_pdf":"http://arxiv.org/pdf/1805.09991v1.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":"lifelong-domain-word-embedding-via-meta","repo_url":"https://github.com/howardhsu/L-DEM","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"lifelong-learning","task_name":"Lifelong learning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.09991","atlas_url":"https://app.syntology.ai/?focus=1805.09991","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}