{"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/thailmcut-unsupervised-pretraining-for-thai","title":"ThaiLMCut: Unsupervised Pretraining for Thai Word Segmentation","arxiv_id":null,"date":"2020-05-01","proceeding":"LREC 2020 5","authors":["Suteera Seeha","Ivan Bilan","Liliana Mamani Sanchez","Johannes Huber","Michael Matuschek","Hinrich Sch{\\\"u}tze"],"abstract":"We propose ThaiLMCut, a semi-supervised approach for Thai word segmentation which utilizes a bi-directional character language model (LM) as a way to leverage useful linguistic knowledge from unlabeled data. After the language model is trained on substantial unlabeled corpora, the weights of its embedding and recurrent layers are transferred to a supervised word segmentation model which continues fine-tuning them on a word segmentation task. Our experimental results demonstrate that applying the LM always leads to a performance gain, especially when the amount of labeled data is small. In such cases, the F1 Score increased by up to 2.02{\\%}. Even on abig labeled dataset, a small improvement gain can still be obtained. The approach has also shown to be very beneficial for out-of-domain settings with a gain in F1 Score of up to 3.13{\\%}. Finally, we show that ThaiLMCut can outperform other open source state-of-the-art models achieving an F1 Score of 98.78{\\%} on the standard benchmark, InterBEST2009.","url_abs":"https://aclanthology.org/2020.lrec-1.858","url_pdf":"https://aclanthology.org/2020.lrec-1.858.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":"thailmcut-unsupervised-pretraining-for-thai","repo_url":"https://github.com/meanna/ThaiLMCUT","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"thai-word-segmentation","task_name":"Thai Word Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/thai-word-tokenization-on-best-2010","task":"Thai Word Segmentation","dataset":"BEST-2010","model":"ThaiLMCut","rank_in_archive_order":3,"of":5,"metrics":{"F1-Score":"0.9878"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}