{"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/mutual-information-and-diverse-decoding","title":"Mutual Information and Diverse Decoding Improve Neural Machine Translation","arxiv_id":"1601.00372","date":"2016-01-04","proceeding":null,"authors":["Jiwei Li","Dan Jurafsky"],"abstract":"Sequence-to-sequence neural translation models learn semantic and syntactic\nrelations between sentence pairs by optimizing the likelihood of the target\ngiven the source, i.e., $p(y|x)$, an objective that ignores other potentially\nuseful sources of information. We introduce an alternative objective function\nfor neural MT that maximizes the mutual information between the source and\ntarget sentences, modeling the bi-directional dependency of sources and\ntargets. We implement the model with a simple re-ranking method, and also\nintroduce a decoding algorithm that increases diversity in the N-best list\nproduced by the first pass. Applied to the WMT German/English and\nFrench/English tasks, the proposed models offers a consistent performance boost\non both standard LSTM and attention-based neural MT architectures.","url_abs":"http://arxiv.org/abs/1601.00372v2","url_pdf":"http://arxiv.org/pdf/1601.00372v2.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":"mutual-information-and-diverse-decoding","repo_url":"https://github.com/hsgodhia/hred","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"re-ranking","task_name":"Re-Ranking"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1601.00372","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}