{"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/transfer-learning-for-low-resource-neural","title":"Transfer Learning for Low-Resource Neural Machine Translation","arxiv_id":"1604.02201","date":"2016-04-08","proceeding":"EMNLP 2016 11","authors":["Barret Zoph","Deniz Yuret","Jonathan May","Kevin Knight"],"abstract":"The encoder-decoder framework for neural machine translation (NMT) has been\nshown effective in large data scenarios, but is much less effective for\nlow-resource languages. We present a transfer learning method that\nsignificantly improves Bleu scores across a range of low-resource languages.\nOur key idea is to first train a high-resource language pair (the parent\nmodel), then transfer some of the learned parameters to the low-resource pair\n(the child model) to initialize and constrain training. Using our transfer\nlearning method we improve baseline NMT models by an average of 5.6 Bleu on\nfour low-resource language pairs. Ensembling and unknown word replacement add\nanother 2 Bleu which brings the NMT performance on low-resource machine\ntranslation close to a strong syntax based machine translation (SBMT) system,\nexceeding its performance on one language pair. Additionally, using the\ntransfer learning model for re-scoring, we can improve the SBMT system by an\naverage of 1.3 Bleu, improving the state-of-the-art on low-resource machine\ntranslation.","url_abs":"http://arxiv.org/abs/1604.02201v1","url_pdf":"http://arxiv.org/pdf/1604.02201v1.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":"transfer-learning-for-low-resource-neural","repo_url":"https://github.com/isi-nlp/Zoph_RNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"low-resource-neural-machine-translation-2","task_name":"Low Resource Neural Machine Translation"},{"task_slug":"low-resource-neural-machine-translation","task_name":"Low-Resource Neural Machine Translation"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1604.02201","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}