{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4VJY6RRVX4BXTYNXMZF43JM7JN","short_pith_number":"pith:4VJY6RRV","schema_version":"1.0","canonical_sha256":"e5538f4635bf0379e1b7664bcda59f4b66034b8cb181f81c3cd3bb3de577da81","source":{"kind":"arxiv","id":"2105.02855","version":2},"attestation_state":"computed","paper":{"title":"Adapting Monolingual Models: Data can be Scarce when Language Similarity is High","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Malvina Nissim, Martijn Bartelds, Martijn Wieling, Wietse de Vries","submitted_at":"2021-05-06T17:43:40Z","abstract_excerpt":"For many (minority) languages, the resources needed to train large models are not available. We investigate the performance of zero-shot transfer learning with as little data as possible, and the influence of language similarity in this process. We retrain the lexical layers of four BERT-based models using data from two low-resource target language varieties, while the Transformer layers are independently fine-tuned on a POS-tagging task in the model's source language. By combining the new lexical layers and fine-tuned Transformer layers, we achieve high task performance for both target langua"},"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":"2105.02855","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-05-06T17:43:40Z","cross_cats_sorted":[],"title_canon_sha256":"da0b8c984f2c333d5e6e0a5aedb6cfa0d64c5917dceea22a987594e322bbcc4d","abstract_canon_sha256":"6a6a26d56b4c9a3202a87b1e9f16b9c3ca92d81c5db63a1dab575623934a083e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:02:15.819628Z","signature_b64":"g3BmPs1NKRqWOXSa0rDSCY3VNdwdIjNmhj9I5I5pUFjB7U9lYbT0K2wDeRunT1RAfqe7mNVlEBb4tSvuuShPAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5538f4635bf0379e1b7664bcda59f4b66034b8cb181f81c3cd3bb3de577da81","last_reissued_at":"2026-07-05T03:02:15.819192Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:02:15.819192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapting Monolingual Models: Data can be Scarce when Language Similarity is High","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Malvina Nissim, Martijn Bartelds, Martijn Wieling, Wietse de Vries","submitted_at":"2021-05-06T17:43:40Z","abstract_excerpt":"For many (minority) languages, the resources needed to train large models are not available. We investigate the performance of zero-shot transfer learning with as little data as possible, and the influence of language similarity in this process. We retrain the lexical layers of four BERT-based models using data from two low-resource target language varieties, while the Transformer layers are independently fine-tuned on a POS-tagging task in the model's source language. By combining the new lexical layers and fine-tuned Transformer layers, we achieve high task performance for both target langua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02855","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2105.02855/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":"2105.02855","created_at":"2026-07-05T03:02:15.819254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.02855v2","created_at":"2026-07-05T03:02:15.819254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02855","created_at":"2026-07-05T03:02:15.819254+00:00"},{"alias_kind":"pith_short_12","alias_value":"4VJY6RRVX4BX","created_at":"2026-07-05T03:02:15.819254+00:00"},{"alias_kind":"pith_short_16","alias_value":"4VJY6RRVX4BXTYNX","created_at":"2026-07-05T03:02:15.819254+00:00"},{"alias_kind":"pith_short_8","alias_value":"4VJY6RRV","created_at":"2026-07-05T03:02:15.819254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.14297","citing_title":"Cross-Lingual Optimization for Language Transfer in Large Language Models","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN","json":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN.json","graph_json":"https://pith.science/api/pith-number/4VJY6RRVX4BXTYNXMZF43JM7JN/graph.json","events_json":"https://pith.science/api/pith-number/4VJY6RRVX4BXTYNXMZF43JM7JN/events.json","paper":"https://pith.science/paper/4VJY6RRV"},"agent_actions":{"view_html":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN","download_json":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN.json","view_paper":"https://pith.science/paper/4VJY6RRV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.02855&json=true","fetch_graph":"https://pith.science/api/pith-number/4VJY6RRVX4BXTYNXMZF43JM7JN/graph.json","fetch_events":"https://pith.science/api/pith-number/4VJY6RRVX4BXTYNXMZF43JM7JN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN/action/storage_attestation","attest_author":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN/action/author_attestation","sign_citation":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN/action/citation_signature","submit_replication":"https://pith.science/pith/4VJY6RRVX4BXTYNXMZF43JM7JN/action/replication_record"}},"created_at":"2026-07-05T03:02:15.819254+00:00","updated_at":"2026-07-05T03:02:15.819254+00:00"}