{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:JK436RQYZQDIBELMVEXKZKCE3T","short_pith_number":"pith:JK436RQY","schema_version":"1.0","canonical_sha256":"4ab9bf4618cc0680916ca92eaca844dcf75d7601298b3462ed9ed242506cd7cb","source":{"kind":"arxiv","id":"1710.11041","version":2},"attestation_state":"computed","paper":{"title":"Unsupervised Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Eneko Agirre, Gorka Labaka, Kyunghyun Cho, Mikel Artetxe","submitted_at":"2017-10-30T16:17:34Z","abstract_excerpt":"In spite of the recent success of neural machine translation (NMT) in standard benchmarks, the lack of large parallel corpora poses a major practical problem for many language pairs. There have been several proposals to alleviate this issue with, for instance, triangulation and semi-supervised learning techniques, but they still require a strong cross-lingual signal. In this work, we completely remove the need of parallel data and propose a novel method to train an NMT system in a completely unsupervised manner, relying on nothing but monolingual corpora. Our model builds upon the recent work "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1710.11041","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-30T16:17:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2349fb24aef19ed3cfd04c2fb4d0633ac68e869652f6432137a5cab351f09ee4","abstract_canon_sha256":"5347bf50ab68c02d47c7e138904cd9fd3eff66a2cd5e6f8fc10c295e0988caef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:22:39.321827Z","signature_b64":"jjC7S7GZbUNSBRkuDHVgtQgM+yylaB7yeESt8CUPfLsSMQmN3TmnsoVUo4bJOhUubVsbKiEzNEDMPWd+YqLZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ab9bf4618cc0680916ca92eaca844dcf75d7601298b3462ed9ed242506cd7cb","last_reissued_at":"2026-05-18T00:22:39.321452Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:22:39.321452Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Eneko Agirre, Gorka Labaka, Kyunghyun Cho, Mikel Artetxe","submitted_at":"2017-10-30T16:17:34Z","abstract_excerpt":"In spite of the recent success of neural machine translation (NMT) in standard benchmarks, the lack of large parallel corpora poses a major practical problem for many language pairs. There have been several proposals to alleviate this issue with, for instance, triangulation and semi-supervised learning techniques, but they still require a strong cross-lingual signal. In this work, we completely remove the need of parallel data and propose a novel method to train an NMT system in a completely unsupervised manner, relying on nothing but monolingual corpora. Our model builds upon the recent work "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.11041","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1710.11041","created_at":"2026-05-18T00:22:39.321505+00:00"},{"alias_kind":"arxiv_version","alias_value":"1710.11041v2","created_at":"2026-05-18T00:22:39.321505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.11041","created_at":"2026-05-18T00:22:39.321505+00:00"},{"alias_kind":"pith_short_12","alias_value":"JK436RQYZQDI","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_16","alias_value":"JK436RQYZQDIBELM","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_8","alias_value":"JK436RQY","created_at":"2026-05-18T12:31:24.725408+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1906.08885","citing_title":"Low-Resource Corpus Filtering using Multilingual Sentence Embeddings","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2009.10297","citing_title":"CodeBLEU: a Method for Automatic Evaluation of Code Synthesis","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"1808.06226","citing_title":"SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T","json":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T.json","graph_json":"https://pith.science/api/pith-number/JK436RQYZQDIBELMVEXKZKCE3T/graph.json","events_json":"https://pith.science/api/pith-number/JK436RQYZQDIBELMVEXKZKCE3T/events.json","paper":"https://pith.science/paper/JK436RQY"},"agent_actions":{"view_html":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T","download_json":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T.json","view_paper":"https://pith.science/paper/JK436RQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1710.11041&json=true","fetch_graph":"https://pith.science/api/pith-number/JK436RQYZQDIBELMVEXKZKCE3T/graph.json","fetch_events":"https://pith.science/api/pith-number/JK436RQYZQDIBELMVEXKZKCE3T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T/action/storage_attestation","attest_author":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T/action/author_attestation","sign_citation":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T/action/citation_signature","submit_replication":"https://pith.science/pith/JK436RQYZQDIBELMVEXKZKCE3T/action/replication_record"}},"created_at":"2026-05-18T00:22:39.321505+00:00","updated_at":"2026-05-18T00:22:39.321505+00:00"}