{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AKQAD7QQWZWDX7YGY3KCLIGR2X","short_pith_number":"pith:AKQAD7QQ","schema_version":"1.0","canonical_sha256":"02a001fe10b66c3bff06c6d425a0d1d5fa31a1c27b3254bac9ec84ef203e57e8","source":{"kind":"arxiv","id":"2412.13922","version":1},"attestation_state":"computed","paper":{"title":"Pipeline Analysis for Developing Instruct LLMs in Low-Resource Languages: A Case Study on Basque","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ander Corral, Ixak Sarasua, Xabier Saralegi","submitted_at":"2024-12-18T15:05:59Z","abstract_excerpt":"Large language models (LLMs) are typically optimized for resource-rich languages like English, exacerbating the gap between high-resource and underrepresented languages. This work presents a detailed analysis of strategies for developing a model capable of following instructions in a low-resource language, specifically Basque, by focusing on three key stages: pre-training, instruction tuning, and alignment with human preferences. Our findings demonstrate that continual pre-training with a high-quality Basque corpus of around 600 million words improves natural language understanding (NLU) of th"},"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":"2412.13922","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-18T15:05:59Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"588f2e663c0bbb569dc2e9613e6d1e38bc7f9b6147a705ac37323e34c3a7ece7","abstract_canon_sha256":"432b19cee52a3cc715af62555d3988dda7b0101dbe95cf436e00ecfd5b88daf9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:51:10.883717Z","signature_b64":"j3VKE3RgP+b+YCnpXJ/lCG4z1eczFADMBzNyf5lcDfd/+oi/Z4I+i6fUQOUZVfa/a32Tp1N1tkU3K3Dx3BBUCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02a001fe10b66c3bff06c6d425a0d1d5fa31a1c27b3254bac9ec84ef203e57e8","last_reissued_at":"2026-07-05T09:51:10.883160Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:51:10.883160Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pipeline Analysis for Developing Instruct LLMs in Low-Resource Languages: A Case Study on Basque","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ander Corral, Ixak Sarasua, Xabier Saralegi","submitted_at":"2024-12-18T15:05:59Z","abstract_excerpt":"Large language models (LLMs) are typically optimized for resource-rich languages like English, exacerbating the gap between high-resource and underrepresented languages. This work presents a detailed analysis of strategies for developing a model capable of following instructions in a low-resource language, specifically Basque, by focusing on three key stages: pre-training, instruction tuning, and alignment with human preferences. Our findings demonstrate that continual pre-training with a high-quality Basque corpus of around 600 million words improves natural language understanding (NLU) of th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13922","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/2412.13922/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":"2412.13922","created_at":"2026-07-05T09:51:10.883226+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.13922v1","created_at":"2026-07-05T09:51:10.883226+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13922","created_at":"2026-07-05T09:51:10.883226+00:00"},{"alias_kind":"pith_short_12","alias_value":"AKQAD7QQWZWD","created_at":"2026-07-05T09:51:10.883226+00:00"},{"alias_kind":"pith_short_16","alias_value":"AKQAD7QQWZWDX7YG","created_at":"2026-07-05T09:51:10.883226+00:00"},{"alias_kind":"pith_short_8","alias_value":"AKQAD7QQ","created_at":"2026-07-05T09:51:10.883226+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X","json":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X.json","graph_json":"https://pith.science/api/pith-number/AKQAD7QQWZWDX7YGY3KCLIGR2X/graph.json","events_json":"https://pith.science/api/pith-number/AKQAD7QQWZWDX7YGY3KCLIGR2X/events.json","paper":"https://pith.science/paper/AKQAD7QQ"},"agent_actions":{"view_html":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X","download_json":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X.json","view_paper":"https://pith.science/paper/AKQAD7QQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.13922&json=true","fetch_graph":"https://pith.science/api/pith-number/AKQAD7QQWZWDX7YGY3KCLIGR2X/graph.json","fetch_events":"https://pith.science/api/pith-number/AKQAD7QQWZWDX7YGY3KCLIGR2X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X/action/storage_attestation","attest_author":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X/action/author_attestation","sign_citation":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X/action/citation_signature","submit_replication":"https://pith.science/pith/AKQAD7QQWZWDX7YGY3KCLIGR2X/action/replication_record"}},"created_at":"2026-07-05T09:51:10.883226+00:00","updated_at":"2026-07-05T09:51:10.883226+00:00"}