{"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/agreement-based-joint-training-for","title":"Agreement-based Joint Training for Bidirectional Attention-based Neural Machine Translation","arxiv_id":"1512.04650","date":"2015-12-15","proceeding":null,"authors":["Yong Cheng","Shiqi Shen","Zhongjun He","wei he","Hua Wu","Maosong Sun","Yang Liu"],"abstract":"The attentional mechanism has proven to be effective in improving end-to-end\nneural machine translation. However, due to the intricate structural divergence\nbetween natural languages, unidirectional attention-based models might only\ncapture partial aspects of attentional regularities. We propose agreement-based\njoint training for bidirectional attention-based end-to-end neural machine\ntranslation. Instead of training source-to-target and target-to-source\ntranslation models independently,our approach encourages the two complementary\nmodels to agree on word alignment matrices on the same training data.\nExperiments on Chinese-English and English-French translation tasks show that\nagreement-based joint training significantly improves both alignment and\ntranslation quality over independent training.","url_abs":"http://arxiv.org/abs/1512.04650v2","url_pdf":"http://arxiv.org/pdf/1512.04650v2.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":"agreement-based-joint-training-for","repo_url":"https://github.com/bagequan/tencent-transformer-with-disagreement","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"word-alignment","task_name":"Word Alignment"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1512.04650","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}