{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ","short_pith_number":"pith:CFHYMP7Z","schema_version":"1.0","canonical_sha256":"114f863ff99c6dca9903cf80a6da1c9e5e47fc027c411dda6442da9f31021863","source":{"kind":"arxiv","id":"2106.16171","version":1},"attestation_state":"computed","paper":{"title":"Revisiting the Primacy of English in Zero-shot Cross-lingual Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Iulia Turc, Jacob Eisenstein, Kenton Lee, Kristina Toutanova, Ming-Wei Chang","submitted_at":"2021-06-30T16:05:57Z","abstract_excerpt":"Despite their success, large pre-trained multilingual models have not completely alleviated the need for labeled data, which is cumbersome to collect for all target languages. Zero-shot cross-lingual transfer is emerging as a practical solution: pre-trained models later fine-tuned on one transfer language exhibit surprising performance when tested on many target languages. English is the dominant source language for transfer, as reinforced by popular zero-shot benchmarks. However, this default choice has not been systematically vetted. In our study, we compare English against other transfer 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":"2106.16171","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-06-30T16:05:57Z","cross_cats_sorted":[],"title_canon_sha256":"b21a0023d1cd094c005ad4aac4696f7586b3c467d54690c2e6f706417f6c794c","abstract_canon_sha256":"833faa8f965600ebdd70c27c4c9b780b57c0af77a3a90ab472f6293840bb0900"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:54:04.197888Z","signature_b64":"EvQ7ZwwJLQTmy8Tpcyvd7CRhz7aUVW8fLEwZH6LON7FzFOmhF0+4T9dTHYWdcEtBCFF6U87aGn+h8QTsxfWuDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"114f863ff99c6dca9903cf80a6da1c9e5e47fc027c411dda6442da9f31021863","last_reissued_at":"2026-07-05T02:54:04.197499Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:54:04.197499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Revisiting the Primacy of English in Zero-shot Cross-lingual Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Iulia Turc, Jacob Eisenstein, Kenton Lee, Kristina Toutanova, Ming-Wei Chang","submitted_at":"2021-06-30T16:05:57Z","abstract_excerpt":"Despite their success, large pre-trained multilingual models have not completely alleviated the need for labeled data, which is cumbersome to collect for all target languages. Zero-shot cross-lingual transfer is emerging as a practical solution: pre-trained models later fine-tuned on one transfer language exhibit surprising performance when tested on many target languages. English is the dominant source language for transfer, as reinforced by popular zero-shot benchmarks. However, this default choice has not been systematically vetted. In our study, we compare English against other transfer la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.16171","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2106.16171/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2106.16171","created_at":"2026-07-05T02:54:04.197560+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.16171v1","created_at":"2026-07-05T02:54:04.197560+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.16171","created_at":"2026-07-05T02:54:04.197560+00:00"},{"alias_kind":"pith_short_12","alias_value":"CFHYMP7ZTRW4","created_at":"2026-07-05T02:54:04.197560+00:00"},{"alias_kind":"pith_short_16","alias_value":"CFHYMP7ZTRW4VGID","created_at":"2026-07-05T02:54:04.197560+00:00"},{"alias_kind":"pith_short_8","alias_value":"CFHYMP7Z","created_at":"2026-07-05T02:54:04.197560+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.08785","citing_title":"Zero-shot Cross-lingual Transfer Learning with Multiple Source and Target Languages for Information Extraction: Language Selection and Adversarial Training","ref_index":36,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ","json":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ.json","graph_json":"https://pith.science/api/pith-number/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/graph.json","events_json":"https://pith.science/api/pith-number/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/events.json","paper":"https://pith.science/paper/CFHYMP7Z"},"agent_actions":{"view_html":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ","download_json":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ.json","view_paper":"https://pith.science/paper/CFHYMP7Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.16171&json=true","fetch_graph":"https://pith.science/api/pith-number/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/graph.json","fetch_events":"https://pith.science/api/pith-number/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/action/storage_attestation","attest_author":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/action/author_attestation","sign_citation":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/action/citation_signature","submit_replication":"https://pith.science/pith/CFHYMP7ZTRW4VGIDZ6AKNWQ4TZ/action/replication_record"}},"created_at":"2026-07-05T02:54:04.197560+00:00","updated_at":"2026-07-05T02:54:04.197560+00:00"}