{"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/thumt-an-open-source-toolkit-for-neural","title":"THUMT: An Open Source Toolkit for Neural Machine Translation","arxiv_id":"1706.06415","date":"2017-06-20","proceeding":null,"authors":["Jiacheng Zhang","Yanzhuo Ding","Shiqi Shen","Yong Cheng","Maosong Sun","Huanbo Luan","Yang Liu"],"abstract":"This paper introduces THUMT, an open-source toolkit for neural machine\ntranslation (NMT) developed by the Natural Language Processing Group at\nTsinghua University. THUMT implements the standard attention-based\nencoder-decoder framework on top of Theano and supports three training\ncriteria: maximum likelihood estimation, minimum risk training, and\nsemi-supervised training. It features a visualization tool for displaying the\nrelevance between hidden states in neural networks and contextual words, which\nhelps to analyze the internal workings of NMT. Experiments on Chinese-English\ndatasets show that THUMT using minimum risk training significantly outperforms\nGroundHog, a state-of-the-art toolkit for NMT.","url_abs":"http://arxiv.org/abs/1706.06415v1","url_pdf":"http://arxiv.org/pdf/1706.06415v1.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":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/Glaceon31/Document-Transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/THUNLP-MT/THUMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/insigh/THUMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/thumt/THUMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/thunlp-mt/ckd","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"thumt-an-open-source-toolkit-for-neural","repo_url":"https://github.com/wangmz15/Chinese-Error-Correction-with-THUMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.06415","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}