{"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/improving-cross-lingual-word-embeddings-by","title":"Improving Cross-Lingual Word Embeddings by Meeting in the Middle","arxiv_id":"1808.08780","date":"2018-08-27","proceeding":"EMNLP 2018 10","authors":["Yerai Doval","Jose Camacho-Collados","Luis Espinosa-Anke","Steven Schockaert"],"abstract":"Cross-lingual word embeddings are becoming increasingly important in\nmultilingual NLP. Recently, it has been shown that these embeddings can be\neffectively learned by aligning two disjoint monolingual vector spaces through\nlinear transformations, using no more than a small bilingual dictionary as\nsupervision. In this work, we propose to apply an additional transformation\nafter the initial alignment step, which moves cross-lingual synonyms towards a\nmiddle point between them. By applying this transformation our aim is to obtain\na better cross-lingual integration of the vector spaces. In addition, and\nperhaps surprisingly, the monolingual spaces also improve by this\ntransformation. This is in contrast to the original alignment, which is\ntypically learned such that the structure of the monolingual spaces is\npreserved. Our experiments confirm that the resulting cross-lingual embeddings\noutperform state-of-the-art models in both monolingual and cross-lingual\nevaluation tasks.","url_abs":"http://arxiv.org/abs/1808.08780v1","url_pdf":"http://arxiv.org/pdf/1808.08780v1.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":"improving-cross-lingual-word-embeddings-by","repo_url":"https://github.com/yeraidm/meemi","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"cross-lingual-word-embeddings","task_name":"Cross-Lingual Word Embeddings"},{"task_slug":"multilingual-nlp","task_name":"Multilingual NLP"},{"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=1808.08780","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}