{"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-language-representations-for","title":"Learning Language Representations for Typology Prediction","arxiv_id":"1707.09569","date":"2017-07-29","proceeding":"EMNLP 2017 9","authors":["Chaitanya Malaviya","Graham Neubig","Patrick Littell"],"abstract":"One central mystery of neural NLP is what neural models \"know\" about their\nsubject matter. When a neural machine translation system learns to translate\nfrom one language to another, does it learn the syntax or semantics of the\nlanguages? Can this knowledge be extracted from the system to fill holes in\nhuman scientific knowledge? Existing typological databases contain relatively\nfull feature specifications for only a few hundred languages. Exploiting the\nexistence of parallel texts in more than a thousand languages, we build a\nmassive many-to-one neural machine translation (NMT) system from 1017 languages\ninto English, and use this to predict information missing from typological\ndatabases. Experiments show that the proposed method is able to infer not only\nsyntactic, but also phonological and phonetic inventory features, and improves\nover a baseline that has access to information about the languages' geographic\nand phylogenetic neighbors.","url_abs":"http://arxiv.org/abs/1707.09569v1","url_pdf":"http://arxiv.org/pdf/1707.09569v1.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-language-representations-for","repo_url":"https://github.com/chaitanyamalaviya/lang-reps","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-language-representations-for","repo_url":"https://github.com/antonisa/lang2vec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.09569","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}