{"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/word-translation-without-parallel-data","title":"Word Translation Without Parallel Data","arxiv_id":"1710.04087","date":"2017-10-11","proceeding":"ICLR 2018 1","authors":["Alexis Conneau","Guillaume Lample","Marc'Aurelio Ranzato","Ludovic Denoyer","Hervé Jégou"],"abstract":"State-of-the-art methods for learning cross-lingual word embeddings have\nrelied on bilingual dictionaries or parallel corpora. Recent studies showed\nthat the need for parallel data supervision can be alleviated with\ncharacter-level information. While these methods showed encouraging results,\nthey are not on par with their supervised counterparts and are limited to pairs\nof languages sharing a common alphabet. In this work, we show that we can build\na bilingual dictionary between two languages without using any parallel\ncorpora, by aligning monolingual word embedding spaces in an unsupervised way.\nWithout using any character information, our model even outperforms existing\nsupervised methods on cross-lingual tasks for some language pairs. Our\nexperiments demonstrate that our method works very well also for distant\nlanguage pairs, like English-Russian or English-Chinese. We finally describe\nexperiments on the English-Esperanto low-resource language pair, on which there\nonly exists a limited amount of parallel data, to show the potential impact of\nour method in fully unsupervised machine translation. Our code, embeddings and\ndictionaries are publicly available.","url_abs":"http://arxiv.org/abs/1710.04087v3","url_pdf":"http://arxiv.org/pdf/1710.04087v3.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":"word-translation-without-parallel-data","repo_url":"https://github.com/facebookresearch/MUSE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/AlexMoreo/funnelling","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","repo_url":"https://github.com/ZahinAkram/Unsupervised-Neural-Machine-Translation-Bangla-To-English-English-to-Bangla-","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/andreapdr/gfun","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","repo_url":"https://github.com/baidu-research/HNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/barnerwothers/MUSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/beinborn/SemanticDrift","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","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":"word-translation-without-parallel-data","repo_url":"https://github.com/sajidahmed12/CSE495-NLP-Assignment-Unsupervised-Neural-Machine-Translation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"word-translation-without-parallel-data","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"}},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/yunsukim86/wbw-lm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/zzzxiaohong/icbt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"word-translation-without-parallel-data","repo_url":"https://github.com/Varchita-Beena/Word-Translation-Without-Parallel-Data","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"cross-lingual-word-embeddings","task_name":"Cross-Lingual Word Embeddings"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"unsupervised-machine-translation","task_name":"Unsupervised Machine Translation"},{"task_slug":"word-alignment","task_name":"Word Alignment"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"},{"task_slug":"word-translation","task_name":"Word Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/word-alignment-on-en-es","task":"Word Alignment","dataset":"en-es","model":"Adv - Refine - CSLS","rank_in_archive_order":2,"of":2,"metrics":{"P@1":"81.7"},"uses_additional_data":false},{"leaderboard":"/sota/word-alignment-on-en-fr","task":"Word Alignment","dataset":"en-fr","model":"Adv - Refine - CSLS","rank_in_archive_order":2,"of":2,"metrics":{"P@1":"82.3"},"uses_additional_data":false},{"leaderboard":"/sota/word-alignment-on-es-en","task":"Word Alignment","dataset":"es-en","model":"Adv - Refine - CSLS","rank_in_archive_order":2,"of":2,"metrics":{"P@1":"83.3"},"uses_additional_data":false},{"leaderboard":"/sota/word-alignment-on-fr-en","task":"Word Alignment","dataset":"fr-en","model":"Adv - Refine - CSLS","rank_in_archive_order":2,"of":2,"metrics":{"P@1":"82.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1710.04087","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1710.04087"}},"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/labdac/charlacompling","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Varchita-Beena/Word-Translation-Without-Parallel-Data","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/yunsukim86/wbw-lm","reach":null},{"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/ZahinAkram/Unsupervised-Neural-Machine-Translation-Bangla-To-English-English-to-Bangla-","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/AlexMoreo/funnelling","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/beinborn/SemanticDrift","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/andreapdr/gfun","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sajidahmed12/CSE495-NLP-Assignment-Unsupervised-Neural-Machine-Translation","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/baidu-research/HNN","reach":{"status":"ok"}},{"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/zzzxiaohong/icbt","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":3,"ran_fixture":1,"ran_violates":2,"unverified":2},"by_repo_kind":{"listed":{"samples":5,"ran":3,"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":8,"samples":[{"code_sha256_prefix":"40cdf0f2f7b382ee","entry":"chunk","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"40cdf0f2f7b382ee"}},{"code_sha256_prefix":"43adfe189ff205a7","entry":"get_candidates","repo":"yunsukim86/wbw-lm","repo_kind":"listed","path":"src/dico_builder.py","file_url":"https://github.com/yunsukim86/wbw-lm/blob/HEAD/src/dico_builder.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"43adfe189ff205a7"}},{"code_sha256_prefix":"a1dac87bef6453c3","entry":"init_with_browne","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"a1dac87bef6453c3"}},{"code_sha256_prefix":"5beb538fcaf76383","entry":"is_zip","repo":"yunsukim86/wbw-lm","repo_kind":"listed","path":"translate.py","file_url":"https://github.com/yunsukim86/wbw-lm/blob/HEAD/translate.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5beb538fcaf76383"}},{"code_sha256_prefix":"9f9b17d2052bad0d","entry":"nll_FA","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"9f9b17d2052bad0d"}},{"code_sha256_prefix":"0e698708dcb2aac8","entry":"open_file","repo":"yunsukim86/wbw-lm","repo_kind":"listed","path":"translate.py","file_url":"https://github.com/yunsukim86/wbw-lm/blob/HEAD/translate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"0e698708dcb2aac8"}},{"code_sha256_prefix":"4ed787d9444bb493","entry":"build_dictionary","repo":"yunsukim86/wbw-lm","repo_kind":"listed","path":"src/dico_builder.py","file_url":"https://github.com/yunsukim86/wbw-lm/blob/HEAD/src/dico_builder.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":"4ed787d9444bb493"}},{"code_sha256_prefix":"052e782956f7ae75","entry":"load_input","repo":"yunsukim86/wbw-lm","repo_kind":"listed","path":"translate.py","file_url":"https://github.com/yunsukim86/wbw-lm/blob/HEAD/translate.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":"052e782956f7ae75"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}