{"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/bootstrapping-transliteration-with","title":"Bootstrapping Transliteration with Constrained Discovery for Low-Resource Languages","arxiv_id":"1809.07807","date":"2018-09-20","proceeding":"EMNLP 2018 10","authors":["Shyam Upadhyay","Jordan Kodner","Dan Roth"],"abstract":"Generating the English transliteration of a name written in a foreign script\nis an important and challenging step in multilingual knowledge acquisition and\ninformation extraction. Existing approaches to transliteration generation\nrequire a large (>5000) number of training examples. This difficulty contrasts\nwith transliteration discovery, a somewhat easier task that involves picking a\nplausible transliteration from a given list. In this work, we present a\nbootstrapping algorithm that uses constrained discovery to improve generation,\nand can be used with as few as 500 training examples, which we show can be\nsourced from annotators in a matter of hours. This opens the task to languages\nfor which large number of training examples are unavailable. We evaluate\ntransliteration generation performance itself, as well the improvement it\nbrings to cross-lingual candidate generation for entity linking, a typical\ndownstream task. We present a comprehensive evaluation of our approach on nine\nlanguages, each written in a unique script.","url_abs":"http://arxiv.org/abs/1809.07807v1","url_pdf":"http://arxiv.org/pdf/1809.07807v1.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":"bootstrapping-transliteration-with","repo_url":"https://github.com/shyamupa/hma-translit","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"entity-linking","task_name":"Entity Linking"},{"task_slug":"transliteration","task_name":"Transliteration"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.07807","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}