{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WXXRBJ3WOFDL3LLUVBI6G72CC2","short_pith_number":"pith:WXXRBJ3W","schema_version":"1.0","canonical_sha256":"b5ef10a7767146bdad74a851e37f4216b63f8e22442650e500c610e685c0751e","source":{"kind":"arxiv","id":"2408.17024","version":2},"attestation_state":"computed","paper":{"title":"InkubaLM: A small language model for low-resource African languages","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anuoluwapo Aremu, Atnafu Lambebo Tonja, Benjamin Rosman, Bonaventure F. P. Dossou, Eric Peter Wairagala, Fadel Thior, Jade Abbott, Jenalea Rajab, Jessica Ojo, Pelonomi Moiloa, Vukosi Marivate","submitted_at":"2024-08-30T05:42:31Z","abstract_excerpt":"High-resource language models often fall short in the African context, where there is a critical need for models that are efficient, accessible, and locally relevant, even amidst significant computing and data constraints. This paper introduces InkubaLM, a small language model with 0.4 billion parameters, which achieves performance comparable to models with significantly larger parameter counts and more extensive training data on tasks such as machine translation, question-answering, AfriMMLU, and the AfriXnli task. Notably, InkubaLM outperforms many larger models in sentiment analysis and dem"},"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":"2408.17024","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-30T05:42:31Z","cross_cats_sorted":[],"title_canon_sha256":"1b91567b4b2f11882b276244afc7abd3654e1c29714f5806fa4f38b416a48f2c","abstract_canon_sha256":"3ebf92212a6d555b8c79bb98597054492a2e54911f6318ec5d4312626ab71a1a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:02:25.529581Z","signature_b64":"jJrulrUDiy4LakvKXVuhXMAOWFk3GirKDUxo2Z2u6IGOLFu33GnMJA/Qbnbja3dKBDJegWFdgcBUv8NveQRXCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b5ef10a7767146bdad74a851e37f4216b63f8e22442650e500c610e685c0751e","last_reissued_at":"2026-07-05T09:02:25.529092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:02:25.529092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"InkubaLM: A small language model for low-resource African languages","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anuoluwapo Aremu, Atnafu Lambebo Tonja, Benjamin Rosman, Bonaventure F. P. Dossou, Eric Peter Wairagala, Fadel Thior, Jade Abbott, Jenalea Rajab, Jessica Ojo, Pelonomi Moiloa, Vukosi Marivate","submitted_at":"2024-08-30T05:42:31Z","abstract_excerpt":"High-resource language models often fall short in the African context, where there is a critical need for models that are efficient, accessible, and locally relevant, even amidst significant computing and data constraints. This paper introduces InkubaLM, a small language model with 0.4 billion parameters, which achieves performance comparable to models with significantly larger parameter counts and more extensive training data on tasks such as machine translation, question-answering, AfriMMLU, and the AfriXnli task. Notably, InkubaLM outperforms many larger models in sentiment analysis and dem"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.17024","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/2408.17024/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":"2408.17024","created_at":"2026-07-05T09:02:25.529146+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.17024v2","created_at":"2026-07-05T09:02:25.529146+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.17024","created_at":"2026-07-05T09:02:25.529146+00:00"},{"alias_kind":"pith_short_12","alias_value":"WXXRBJ3WOFDL","created_at":"2026-07-05T09:02:25.529146+00:00"},{"alias_kind":"pith_short_16","alias_value":"WXXRBJ3WOFDL3LLU","created_at":"2026-07-05T09:02:25.529146+00:00"},{"alias_kind":"pith_short_8","alias_value":"WXXRBJ3W","created_at":"2026-07-05T09:02:25.529146+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22274","citing_title":"From Speech to Text Corpora: Evaluating ASR-Based Data Acquisition for Low-Resource Fongbe and Hausa","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22828","citing_title":"A Survey of Text and Speech Resources for Hausa and Fongbe: Availability, Quality, and Gaps for NLP Development","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2412.04497","citing_title":"Opportunities and Challenges of Large Language Models for Low-Resource Languages in Humanities Research","ref_index":132,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2","json":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2.json","graph_json":"https://pith.science/api/pith-number/WXXRBJ3WOFDL3LLUVBI6G72CC2/graph.json","events_json":"https://pith.science/api/pith-number/WXXRBJ3WOFDL3LLUVBI6G72CC2/events.json","paper":"https://pith.science/paper/WXXRBJ3W"},"agent_actions":{"view_html":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2","download_json":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2.json","view_paper":"https://pith.science/paper/WXXRBJ3W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.17024&json=true","fetch_graph":"https://pith.science/api/pith-number/WXXRBJ3WOFDL3LLUVBI6G72CC2/graph.json","fetch_events":"https://pith.science/api/pith-number/WXXRBJ3WOFDL3LLUVBI6G72CC2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2/action/storage_attestation","attest_author":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2/action/author_attestation","sign_citation":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2/action/citation_signature","submit_replication":"https://pith.science/pith/WXXRBJ3WOFDL3LLUVBI6G72CC2/action/replication_record"}},"created_at":"2026-07-05T09:02:25.529146+00:00","updated_at":"2026-07-05T09:02:25.529146+00:00"}