{"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/a-convolutional-encoder-model-for-neural","title":"A Convolutional Encoder Model for Neural Machine Translation","arxiv_id":"1611.02344","date":"2016-11-07","proceeding":"ACL 2017 7","authors":["Jonas Gehring","Michael Auli","David Grangier","Yann N. Dauphin"],"abstract":"The prevalent approach to neural machine translation relies on bi-directional\nLSTMs to encode the source sentence. In this paper we present a faster and\nsimpler architecture based on a succession of convolutional layers. This allows\nto encode the entire source sentence simultaneously compared to recurrent\nnetworks for which computation is constrained by temporal dependencies. On\nWMT'16 English-Romanian translation we achieve competitive accuracy to the\nstate-of-the-art and we outperform several recently published results on the\nWMT'15 English-German task. Our models obtain almost the same accuracy as a\nvery deep LSTM setup on WMT'14 English-French translation. Our convolutional\nencoder speeds up CPU decoding by more than two times at the same or higher\naccuracy as a strong bi-directional LSTM baseline.","url_abs":"http://arxiv.org/abs/1611.02344v3","url_pdf":"http://arxiv.org/pdf/1611.02344v3.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":"a-convolutional-encoder-model-for-neural","repo_url":"https://github.com/facebookresearch/fairseq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"a-convolutional-encoder-model-for-neural","repo_url":"https://github.com/siyuofzhou/CNNSeqToSeq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"model","task_name":"model"}],"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":[{"leaderboard":"/sota/machine-translation-on-iwslt2015-german","task":"Machine Translation","dataset":"IWSLT2015 German-English","model":"Conv-LSTM (deep+pos)","rank_in_archive_order":7,"of":15,"metrics":{"BLEU score":"30.4"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2014-english-french","task":"Machine Translation","dataset":"WMT2014 English-French","model":"Deep Convolutional Encoder; single-layer decoder","rank_in_archive_order":46,"of":57,"metrics":{"BLEU score":"35.7"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2016-english-1","task":"Machine Translation","dataset":"WMT2016 English-Romanian","model":"Deep Convolutional Encoder; single-layer decoder","rank_in_archive_order":14,"of":21,"metrics":{"BLEU score":"27.8"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2016-english-1","task":"Machine Translation","dataset":"WMT2016 English-Romanian","model":"BiLSTM","rank_in_archive_order":15,"of":21,"metrics":{"BLEU score":"27.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1611.02344","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}