{"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/non-autoregressive-neural-machine-translation-1","title":"Non-Autoregressive Neural Machine Translation","arxiv_id":"1711.02281","date":"2017-11-07","proceeding":"ICLR 2018 1","authors":["Jiatao Gu","James Bradbury","Caiming Xiong","Victor O. K. Li","Richard Socher"],"abstract":"Existing approaches to neural machine translation condition each output word\non previously generated outputs. We introduce a model that avoids this\nautoregressive property and produces its outputs in parallel, allowing an order\nof magnitude lower latency during inference. Through knowledge distillation,\nthe use of input token fertilities as a latent variable, and policy gradient\nfine-tuning, we achieve this at a cost of as little as 2.0 BLEU points relative\nto the autoregressive Transformer network used as a teacher. We demonstrate\nsubstantial cumulative improvements associated with each of the three aspects\nof our training strategy, and validate our approach on IWSLT 2016\nEnglish-German and two WMT language pairs. By sampling fertilities in parallel\nat inference time, our non-autoregressive model achieves near-state-of-the-art\nperformance of 29.8 BLEU on WMT 2016 English-Romanian.","url_abs":"http://arxiv.org/abs/1711.02281v2","url_pdf":"http://arxiv.org/pdf/1711.02281v2.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":"non-autoregressive-neural-machine-translation-1","repo_url":"https://github.com/salesforce/nonauto-nmt","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"non-autoregressive-neural-machine-translation-1","repo_url":"https://github.com/MultiPath/NA-NMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"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":"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":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/machine-translation-on-iwslt2015-english","task":"Machine Translation","dataset":"IWSLT2015 English-German","model":"NAT +FT + NPD","rank_in_archive_order":3,"of":8,"metrics":{"BLEU score":"28.16"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2014-english-german","task":"Machine Translation","dataset":"WMT2014 English-German","model":"NAT +FT + NPD","rank_in_archive_order":82,"of":91,"metrics":{"BLEU score":"19.17"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2014-german-english","task":"Machine Translation","dataset":"WMT2014 German-English","model":"NAT +FT + NPD","rank_in_archive_order":15,"of":16,"metrics":{"BLEU score":"23.20"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2016-english-1","task":"Machine Translation","dataset":"WMT2016 English-Romanian","model":"NAT +FT + NPD","rank_in_archive_order":9,"of":21,"metrics":{"BLEU score":"29.79"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2016-romanian","task":"Machine Translation","dataset":"WMT2016 Romanian-English","model":"NAT +FT + NPD","rank_in_archive_order":16,"of":21,"metrics":{"BLEU score":"31.44"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.02281","atlas_url":"https://app.syntology.ai/?focus=1711.02281","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1711.02281"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/salesforce/nonauto-nmt","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MultiPath/NA-NMT","reach":null}],"summary":{"ran_fixture":1,"ran_draft_wrong":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"listed":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"143671444bc0de03","entry":"cutoff","repo":"salesforce/nonauto-nmt","repo_kind":"official","path":"decode.py","file_url":"https://github.com/salesforce/nonauto-nmt/blob/HEAD/decode.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"143671444bc0de03"}},{"code_sha256_prefix":"c930e24b7e6e73fd","entry":"masked_sort","repo":"MultiPath/NA-NMT","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/MultiPath/NA-NMT/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c930e24b7e6e73fd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}