{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PM442M4ELKUIJYTKA74QWFWJAN","short_pith_number":"pith:PM442M4E","schema_version":"1.0","canonical_sha256":"7b39cd33845aa884e26a07f90b16c9037af59244d523e576636f2c2888450544","source":{"kind":"arxiv","id":"2412.06484","version":2},"attestation_state":"computed","paper":{"title":"Small Languages, Big Models: A Study of Continual Training on Languages of Norway","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrey Kutuzov, David Samuel, Erik Velldal, Lilja {\\O}vrelid, Lucas Georges Gabriel Charpentier, Stephan Oepen, Vladislav Mikhailov","submitted_at":"2024-12-09T13:34:23Z","abstract_excerpt":"Training large language models requires vast amounts of data, posing a challenge for less widely spoken languages like Norwegian and even more so for truly low-resource languages like Northern S\\'ami. To address this issue, we present a novel three-stage continual training approach that substantially improves the downstream performance together with the inference efficiency for the target languages. Based on our findings, we train, evaluate, and openly release a new generative language model for Norwegian Bokm\\r{a}l, Nynorsk, and Northern S\\'ami with 11.4 billion parameters: NorMistral-11B."},"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":"2412.06484","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-09T13:34:23Z","cross_cats_sorted":[],"title_canon_sha256":"3822cbc61125372b7170d49c1ed989adaebaa274db0eb32c617698714f7b3fee","abstract_canon_sha256":"449500cb4b7a344e8fcd92ef597c58982418318a3e02d7be76152986fc451cd0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:25.829710Z","signature_b64":"RJelK4O3f69iBLhrEAG6WapaVS09QZrNkiYUIwqPS6ztAurHGLMAFvM6JGDQ2AgBq/6FUpEXKnS9mvWgfG/OAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b39cd33845aa884e26a07f90b16c9037af59244d523e576636f2c2888450544","last_reissued_at":"2026-07-05T10:08:25.829224Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:25.829224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Small Languages, Big Models: A Study of Continual Training on Languages of Norway","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrey Kutuzov, David Samuel, Erik Velldal, Lilja {\\O}vrelid, Lucas Georges Gabriel Charpentier, Stephan Oepen, Vladislav Mikhailov","submitted_at":"2024-12-09T13:34:23Z","abstract_excerpt":"Training large language models requires vast amounts of data, posing a challenge for less widely spoken languages like Norwegian and even more so for truly low-resource languages like Northern S\\'ami. To address this issue, we present a novel three-stage continual training approach that substantially improves the downstream performance together with the inference efficiency for the target languages. Based on our findings, we train, evaluate, and openly release a new generative language model for Norwegian Bokm\\r{a}l, Nynorsk, and Northern S\\'ami with 11.4 billion parameters: NorMistral-11B."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.06484","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/2412.06484/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":"2412.06484","created_at":"2026-07-05T10:08:25.829289+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.06484v2","created_at":"2026-07-05T10:08:25.829289+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.06484","created_at":"2026-07-05T10:08:25.829289+00:00"},{"alias_kind":"pith_short_12","alias_value":"PM442M4ELKUI","created_at":"2026-07-05T10:08:25.829289+00:00"},{"alias_kind":"pith_short_16","alias_value":"PM442M4ELKUIJYTK","created_at":"2026-07-05T10:08:25.829289+00:00"},{"alias_kind":"pith_short_8","alias_value":"PM442M4E","created_at":"2026-07-05T10:08:25.829289+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.06692","citing_title":"Multi-label Scandinavian Language Identification (SLIDE)","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN","json":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN.json","graph_json":"https://pith.science/api/pith-number/PM442M4ELKUIJYTKA74QWFWJAN/graph.json","events_json":"https://pith.science/api/pith-number/PM442M4ELKUIJYTKA74QWFWJAN/events.json","paper":"https://pith.science/paper/PM442M4E"},"agent_actions":{"view_html":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN","download_json":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN.json","view_paper":"https://pith.science/paper/PM442M4E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.06484&json=true","fetch_graph":"https://pith.science/api/pith-number/PM442M4ELKUIJYTKA74QWFWJAN/graph.json","fetch_events":"https://pith.science/api/pith-number/PM442M4ELKUIJYTKA74QWFWJAN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN/action/storage_attestation","attest_author":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN/action/author_attestation","sign_citation":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN/action/citation_signature","submit_replication":"https://pith.science/pith/PM442M4ELKUIJYTKA74QWFWJAN/action/replication_record"}},"created_at":"2026-07-05T10:08:25.829289+00:00","updated_at":"2026-07-05T10:08:25.829289+00:00"}