{"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/unsupervised-machine-translation-using","title":"Unsupervised Machine Translation Using Monolingual Corpora Only","arxiv_id":"1711.00043","date":"2017-10-31","proceeding":"ICLR 2018 1","authors":["Guillaume Lample","Alexis Conneau","Ludovic Denoyer","Marc'Aurelio Ranzato"],"abstract":"Machine translation has recently achieved impressive performance thanks to\nrecent advances in deep learning and the availability of large-scale parallel\ncorpora. There have been numerous attempts to extend these successes to\nlow-resource language pairs, yet requiring tens of thousands of parallel\nsentences. In this work, we take this research direction to the extreme and\ninvestigate whether it is possible to learn to translate even without any\nparallel data. We propose a model that takes sentences from monolingual corpora\nin two different languages and maps them into the same latent space. By\nlearning to reconstruct in both languages from this shared feature space, the\nmodel effectively learns to translate without using any labeled data. We\ndemonstrate our model on two widely used datasets and two language pairs,\nreporting BLEU scores of 32.8 and 15.1 on the Multi30k and WMT English-French\ndatasets, without using even a single parallel sentence at training time.","url_abs":"http://arxiv.org/abs/1711.00043v2","url_pdf":"http://arxiv.org/pdf/1711.00043v2.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":"unsupervised-machine-translation-using","repo_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/YovaKem/generalized-procrustes-MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/babylonhealth/MultilingualFactorAnalysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/barnerwothers/MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/facebookresearch/MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/freedombenLiu/MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/jiajunhua/facebookresearch-MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/keleog/PidginUNMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/labdac/charlacompling","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/makozi/AfrikaansNMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/maochf/MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/migonch/unsupervised_mt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/sabetAI/bucc-eval","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"unsupervised-machine-translation-using","repo_url":"https://github.com/samnguyen8991/Facebook-MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"unsupervised-machine-translation","task_name":"Unsupervised Machine Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/machine-translation-on-wmt2016-english-german","task":"Machine Translation","dataset":"WMT2016 English-German","model":"Unsupervised S2S with attention","rank_in_archive_order":11,"of":12,"metrics":{"BLEU score":"9.64"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-wmt2016-german-english","task":"Machine Translation","dataset":"WMT2016 German-English","model":"Unsupervised S2S with attention","rank_in_archive_order":7,"of":8,"metrics":{"BLEU score":"13.33"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.00043","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1711.00043"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/makozi/AfrikaansNMT","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/labdac/charlacompling","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/YovaKem/generalized-procrustes-MUSE","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/babylonhealth/MultilingualFactorAnalysis","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/barnerwothers/MUSE","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/maochf/MUSE","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/freedombenLiu/MUSE","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/MUSE","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/migonch/unsupervised_mt","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/samnguyen8991/Facebook-MUSE","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jiajunhua/facebookresearch-MUSE","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sabetAI/bucc-eval","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/keleog/PidginUNMT","reach":null}],"summary":{"ran_draft_wrong":1,"ran_fixture":1,"ran_violates":1,"unverified":11},"by_repo_kind":{"listed":{"samples":14,"ran":3,"repositories":3}},"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":5,"samples":[{"code_sha256_prefix":"40cdf0f2f7b382ee","entry":"chunk","repo":"barnerwothers/MUSE","repo_kind":"listed","path":"src/alignment_functions.py","file_url":"https://github.com/barnerwothers/MUSE/blob/HEAD/src/alignment_functions.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":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"40cdf0f2f7b382ee"}},{"code_sha256_prefix":"a1dac87bef6453c3","entry":"init_with_browne","repo":"barnerwothers/MUSE","repo_kind":"listed","path":"src/alignment_functions.py","file_url":"https://github.com/barnerwothers/MUSE/blob/HEAD/src/alignment_functions.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"a1dac87bef6453c3"}},{"code_sha256_prefix":"9f9b17d2052bad0d","entry":"nll_FA","repo":"barnerwothers/MUSE","repo_kind":"listed","path":"src/alignment_functions.py","file_url":"https://github.com/barnerwothers/MUSE/blob/HEAD/src/alignment_functions.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"9f9b17d2052bad0d"}},{"code_sha256_prefix":"1017986b046fa3e5","entry":"bow","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1017986b046fa3e5"}},{"code_sha256_prefix":"0a6749c257608531","entry":"build_dictionary","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/dico_builder.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/dico_builder.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0a6749c257608531"}},{"code_sha256_prefix":"420c0f482a285b84","entry":"build_model","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/models.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"420c0f482a285b84"}},{"code_sha256_prefix":"095dc5970a2bfba9","entry":"build_seq2seq_model","repo":"keleog/PidginUNMT","repo_kind":"listed","path":"src/model/seq2seq.py","file_url":"https://github.com/keleog/PidginUNMT/blob/HEAD/src/model/seq2seq.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"095dc5970a2bfba9"}},{"code_sha256_prefix":"00b78a626a078a5b","entry":"create_logger","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/logger.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/logger.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"00b78a626a078a5b"}},{"code_sha256_prefix":"069588d6ca04276f","entry":"get_candidates","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/dico_builder.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/dico_builder.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"069588d6ca04276f"}},{"code_sha256_prefix":"32cc6430b851519d","entry":"get_init_state","repo":"keleog/PidginUNMT","repo_kind":"listed","path":"src/model/seq2seq.py","file_url":"https://github.com/keleog/PidginUNMT/blob/HEAD/src/model/seq2seq.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"32cc6430b851519d"}},{"code_sha256_prefix":"21a3d4ec06ede4eb","entry":"initialize_exp","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"21a3d4ec06ede4eb"}},{"code_sha256_prefix":"2b93faa3b4ec4872","entry":"load_fasttext_model","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2b93faa3b4ec4872"}},{"code_sha256_prefix":"b68d645b4c6b0a2a","entry":"read_dict","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"create_embeddings.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/create_embeddings.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b68d645b4c6b0a2a"}},{"code_sha256_prefix":"affe0a63af29c7b1","entry":"sample_language_pair","repo":"Babylonpartners/MultilingualFactorAnalysis","repo_kind":"listed","path":"create_embeddings.py","file_url":"https://github.com/Babylonpartners/MultilingualFactorAnalysis/blob/HEAD/create_embeddings.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"affe0a63af29c7b1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}