{"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/the-best-of-both-worlds-combining-recent","title":"The Best of Both Worlds: Combining Recent Advances in Neural Machine Translation","arxiv_id":"1804.09849","date":"2018-04-26","proceeding":"ACL 2018 7","authors":["Mia Xu Chen","Orhan Firat","Ankur Bapna","Melvin Johnson","Wolfgang Macherey","George Foster","Llion Jones","Niki Parmar","Mike Schuster","Zhifeng Chen","Yonghui Wu","Macduff Hughes"],"abstract":"The past year has witnessed rapid advances in sequence-to-sequence (seq2seq)\nmodeling for Machine Translation (MT). The classic RNN-based approaches to MT\nwere first out-performed by the convolutional seq2seq model, which was then\nout-performed by the more recent Transformer model. Each of these new\napproaches consists of a fundamental architecture accompanied by a set of\nmodeling and training techniques that are in principle applicable to other\nseq2seq architectures. In this paper, we tease apart the new architectures and\ntheir accompanying techniques in two ways. First, we identify several key\nmodeling and training techniques, and apply them to the RNN architecture,\nyielding a new RNMT+ model that outperforms all of the three fundamental\narchitectures on the benchmark WMT'14 English to French and English to German\ntasks. Second, we analyze the properties of each fundamental seq2seq\narchitecture and devise new hybrid architectures intended to combine their\nstrengths. Our hybrid models obtain further improvements, outperforming the\nRNMT+ model on both benchmark datasets.","url_abs":"http://arxiv.org/abs/1804.09849v2","url_pdf":"http://arxiv.org/pdf/1804.09849v2.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":"the-best-of-both-worlds-combining-recent","repo_url":"https://github.com/tensorflow/lingvo","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"the-best-of-both-worlds-combining-recent","repo_url":"https://github.com/duyvuleo/Transformer-DyNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"the-best-of-both-worlds-combining-recent","repo_url":"https://github.com/zysite/post","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"seq2seq","method_name":"Seq2Seq"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/machine-translation-on-wmt2014-english-french","task":"Machine Translation","dataset":"WMT2014 English-French","model":"RNMT+","rank_in_archive_order":27,"of":57,"metrics":{"BLEU score":"41.0","Hardware Burden":"132G","Operations per network pass":"2.81G"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2014-english-german","task":"Machine Translation","dataset":"WMT2014 English-German","model":"RNMT+","rank_in_archive_order":41,"of":91,"metrics":{"BLEU score":"28.5","Hardware Burden":"44G","Operations per network pass":"2.81G"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.09849","atlas_url":"https://app.syntology.ai/?focus=1804.09849","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}