{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PKI6VVWNHGTBT7TQDE3F4GYUW3","short_pith_number":"pith:PKI6VVWN","schema_version":"1.0","canonical_sha256":"7a91ead6cd39a619fe7019365e1b14b6dfb233752d5ce049bf685ca71f31c536","source":{"kind":"arxiv","id":"1906.01181","version":1},"attestation_state":"computed","paper":{"title":"Improved Zero-shot Neural Machine Translation via Ignoring Spurious Correlations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiatao Gu, Kyunghyun Cho, Victor O.K. Li, Yong Wang","submitted_at":"2019-06-04T03:30:22Z","abstract_excerpt":"Zero-shot translation, translating between language pairs on which a Neural Machine Translation (NMT) system has never been trained, is an emergent property when training the system in multilingual settings. However, naive training for zero-shot NMT easily fails, and is sensitive to hyper-parameter setting. The performance typically lags far behind the more conventional pivot-based approach which translates twice using a third language as a pivot. In this work, we address the degeneracy problem due to capturing spurious correlations by quantitatively analyzing the mutual information between la"},"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":"1906.01181","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-04T03:30:22Z","cross_cats_sorted":[],"title_canon_sha256":"c44317b976349e14676e8ce083f8520c6f7d4f5b4ffb007994d2bc1f66974a5a","abstract_canon_sha256":"04f1da7f99b0c809e7054cad3191080419fb5f59d173ef12b96348148e9db323"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:17.665234Z","signature_b64":"6aNZPdpOiTMqlNYkK0XuT+BRlFZqVAYTDJWdeT5of7ZHepP4QCjB3uTdtzG5T6g6TWDchoWm77bYWvvx9phYAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a91ead6cd39a619fe7019365e1b14b6dfb233752d5ce049bf685ca71f31c536","last_reissued_at":"2026-05-17T23:44:17.664514Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:17.664514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved Zero-shot Neural Machine Translation via Ignoring Spurious Correlations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiatao Gu, Kyunghyun Cho, Victor O.K. Li, Yong Wang","submitted_at":"2019-06-04T03:30:22Z","abstract_excerpt":"Zero-shot translation, translating between language pairs on which a Neural Machine Translation (NMT) system has never been trained, is an emergent property when training the system in multilingual settings. However, naive training for zero-shot NMT easily fails, and is sensitive to hyper-parameter setting. The performance typically lags far behind the more conventional pivot-based approach which translates twice using a third language as a pivot. In this work, we address the degeneracy problem due to capturing spurious correlations by quantitatively analyzing the mutual information between la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.01181","kind":"arxiv","version":1},"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":"1906.01181","created_at":"2026-05-17T23:44:17.664625+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.01181v1","created_at":"2026-05-17T23:44:17.664625+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.01181","created_at":"2026-05-17T23:44:17.664625+00:00"},{"alias_kind":"pith_short_12","alias_value":"PKI6VVWNHGTB","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_16","alias_value":"PKI6VVWNHGTBT7TQ","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_8","alias_value":"PKI6VVWN","created_at":"2026-05-18T12:33:24.271573+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1906.09675","citing_title":"Evaluating the Supervised and Zero-shot Performance of Multi-lingual Translation Models","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3","json":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3.json","graph_json":"https://pith.science/api/pith-number/PKI6VVWNHGTBT7TQDE3F4GYUW3/graph.json","events_json":"https://pith.science/api/pith-number/PKI6VVWNHGTBT7TQDE3F4GYUW3/events.json","paper":"https://pith.science/paper/PKI6VVWN"},"agent_actions":{"view_html":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3","download_json":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3.json","view_paper":"https://pith.science/paper/PKI6VVWN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.01181&json=true","fetch_graph":"https://pith.science/api/pith-number/PKI6VVWNHGTBT7TQDE3F4GYUW3/graph.json","fetch_events":"https://pith.science/api/pith-number/PKI6VVWNHGTBT7TQDE3F4GYUW3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3/action/storage_attestation","attest_author":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3/action/author_attestation","sign_citation":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3/action/citation_signature","submit_replication":"https://pith.science/pith/PKI6VVWNHGTBT7TQDE3F4GYUW3/action/replication_record"}},"created_at":"2026-05-17T23:44:17.664625+00:00","updated_at":"2026-05-17T23:44:17.664625+00:00"}