{"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-lemmatization-of-non-standard","title":"Improving Lemmatization of Non-Standard Languages with Joint Learning","arxiv_id":"1903.06939","date":"2019-03-16","proceeding":"NAACL 2019 6","authors":["Enrique Manjavacas","Ákos Kádár","Mike Kestemont"],"abstract":"Lemmatization of standard languages is concerned with (i) abstracting over\nmorphological differences and (ii) resolving token-lemma ambiguities of\ninflected words in order to map them to a dictionary headword. In the present\npaper we aim to improve lemmatization performance on a set of non-standard\nhistorical languages in which the difficulty is increased by an additional\naspect (iii): spelling variation due to lacking orthographic standards. We\napproach lemmatization as a string-transduction task with an encoder-decoder\narchitecture which we enrich with sentence context information using a\nhierarchical sentence encoder. We show significant improvements over the\nstate-of-the-art when training the sentence encoder jointly for lemmatization\nand language modeling. Crucially, our architecture does not require POS or\nmorphological annotations, which are not always available for historical\ncorpora. Additionally, we also test the proposed model on a set of\ntypologically diverse standard languages showing results on par or better than\na model without enhanced sentence representations and previous state-of-the-art\nsystems. Finally, to encourage future work on processing of non-standard\nvarieties, we release the dataset of non-standard languages underlying the\npresent study, based on openly accessible sources.","url_abs":"http://arxiv.org/abs/1903.06939v1","url_pdf":"http://arxiv.org/pdf/1903.06939v1.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-lemmatization-of-non-standard","repo_url":"https://github.com/emanjavacas/pie","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"improving-lemmatization-of-non-standard","repo_url":"https://github.com/emanjavacas/pie-data","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"lemma","task_name":"LEMMA"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"lemmatization","task_name":"Lemmatization"},{"task_slug":"pos","task_name":"POS"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}