{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Y652OHKLRI6OU2ZW22BLVTB5VO","short_pith_number":"pith:Y652OHKL","schema_version":"1.0","canonical_sha256":"c7bba71d4b8a3cea6b36d682bacc3dab9e58bc1dfff000140123b5d7b1495b9d","source":{"kind":"arxiv","id":"2403.06354","version":1},"attestation_state":"computed","paper":{"title":"Amharic LLaMA and LLaVA: Multimodal LLMs for Low Resource Languages","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Michael Andersland","submitted_at":"2024-03-11T01:04:36Z","abstract_excerpt":"Large Language Models (LLMs) like GPT-4 and LLaMA have shown incredible proficiency at natural language processing tasks and have even begun to excel at tasks across other modalities such as vision and audio. Despite their success, LLMs often struggle to perform well on low-resource languages because there is so little training data available. This shortcoming is especially prevalent with open source models. In this work, we explore training LLaMA-2 to speak Amharic, a language which is spoken by over 50 million people world wide, but has orders of magnitude less data available than languages "},"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":"2403.06354","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-11T01:04:36Z","cross_cats_sorted":[],"title_canon_sha256":"fbbc8018d9a2420c87eb5fde0aa015678e0cea587723bb2468ae7eddb236b749","abstract_canon_sha256":"4d1e06f69dd8f679ecbc8854ccc1ba4ffe97e85c123cbd4b272c88d4e87bea4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:22.379461Z","signature_b64":"3ZetNDn7suNiD9bkKgBmeZID0VrTqUYuaK8LixlBiEc8yVheSTrPasElUKF5HNOFRpxvOoeecEgxQQX4LSYfAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7bba71d4b8a3cea6b36d682bacc3dab9e58bc1dfff000140123b5d7b1495b9d","last_reissued_at":"2026-07-05T07:54:22.378984Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:22.378984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Amharic LLaMA and LLaVA: Multimodal LLMs for Low Resource Languages","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Michael Andersland","submitted_at":"2024-03-11T01:04:36Z","abstract_excerpt":"Large Language Models (LLMs) like GPT-4 and LLaMA have shown incredible proficiency at natural language processing tasks and have even begun to excel at tasks across other modalities such as vision and audio. Despite their success, LLMs often struggle to perform well on low-resource languages because there is so little training data available. This shortcoming is especially prevalent with open source models. In this work, we explore training LLaMA-2 to speak Amharic, a language which is spoken by over 50 million people world wide, but has orders of magnitude less data available than languages "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.06354","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/2403.06354/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":"2403.06354","created_at":"2026-07-05T07:54:22.379036+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.06354v1","created_at":"2026-07-05T07:54:22.379036+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.06354","created_at":"2026-07-05T07:54:22.379036+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y652OHKLRI6O","created_at":"2026-07-05T07:54:22.379036+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y652OHKLRI6OU2ZW","created_at":"2026-07-05T07:54:22.379036+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y652OHKL","created_at":"2026-07-05T07:54:22.379036+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.11073","citing_title":"CLAIM: Mitigating Multilingual Object Hallucination in Large Vision-Language Models with Cross-Lingual Attention Intervention","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO","json":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO.json","graph_json":"https://pith.science/api/pith-number/Y652OHKLRI6OU2ZW22BLVTB5VO/graph.json","events_json":"https://pith.science/api/pith-number/Y652OHKLRI6OU2ZW22BLVTB5VO/events.json","paper":"https://pith.science/paper/Y652OHKL"},"agent_actions":{"view_html":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO","download_json":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO.json","view_paper":"https://pith.science/paper/Y652OHKL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.06354&json=true","fetch_graph":"https://pith.science/api/pith-number/Y652OHKLRI6OU2ZW22BLVTB5VO/graph.json","fetch_events":"https://pith.science/api/pith-number/Y652OHKLRI6OU2ZW22BLVTB5VO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO/action/storage_attestation","attest_author":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO/action/author_attestation","sign_citation":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO/action/citation_signature","submit_replication":"https://pith.science/pith/Y652OHKLRI6OU2ZW22BLVTB5VO/action/replication_record"}},"created_at":"2026-07-05T07:54:22.379036+00:00","updated_at":"2026-07-05T07:54:22.379036+00:00"}