{"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/universal-language-model-fine-tuning-with","title":"Universal Language Model Fine-Tuning with Subword Tokenization for Polish","arxiv_id":"1810.10222","date":"2018-10-24","proceeding":null,"authors":["Piotr Czapla","Jeremy Howard","Marcin Kardas"],"abstract":"Universal Language Model for Fine-tuning [arXiv:1801.06146] (ULMFiT) is one\nof the first NLP methods for efficient inductive transfer learning.\nUnsupervised pretraining results in improvements on many NLP tasks for English.\nIn this paper, we describe a new method that uses subword tokenization to adapt\nULMFiT to languages with high inflection. Our approach results in a new\nstate-of-the-art for the Polish language, taking first place in Task 3 of\nPolEval'18. After further training, our final model outperformed the second\nbest model by 35%. We have open-sourced our pretrained models and code.","url_abs":"http://arxiv.org/abs/1810.10222v1","url_pdf":"http://arxiv.org/pdf/1810.10222v1.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":"universal-language-model-fine-tuning-with","repo_url":"https://github.com/simecek/Czech-ULMFiT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"universal-language-model-fine-tuning-with","repo_url":"https://github.com/n-waves/poleval2018","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}