{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:S376LO52B6YUB5VTTRHYSBWNMD","short_pith_number":"pith:S376LO52","schema_version":"1.0","canonical_sha256":"96ffe5bbba0fb140f6b39c4f8906cd60e1023dded137d7d445ebdff8f03ecd9d","source":{"kind":"arxiv","id":"2502.09056","version":3},"attestation_state":"computed","paper":{"title":"Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging -- An Open Recipe","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kasima Tharnpipitchai, Kunat Pipatanakul, Pittawat Taveekitworachai, Potsawee Manakul","submitted_at":"2025-02-13T08:10:45Z","abstract_excerpt":"This paper investigates data selection and model merging methodologies aimed at incorporating advanced reasoning capabilities such as those of DeepSeek R1 into language-specific large language models (LLMs), with a particular focus on the Thai LLM. Our goal is to enhance the reasoning capabilities of language-specific LLMs while maintaining their target language abilities. DeepSeek R1 excels in reasoning but primarily benefits high-resource languages such as English and Chinese. However, low-resource languages remain underserved due to the dominance of English-centric training data and model o"},"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":"2502.09056","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-13T08:10:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"634205d7522c878326d540c8dfda59bc8d34675635a80a7ac63036d32a2ac7bf","abstract_canon_sha256":"b280a9220151b9f32cb0a376f42c56af8b810c34f2043b66e64e0dbf8cdbba5f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:16.934801Z","signature_b64":"3fEvrMLab1uR5rTjQGxvgnMnnpZA7FpIx4lkH11B9GQBLZkbjVV/XdTaW7x/4+rdehkV1JquWGYrmsLq+1oACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"96ffe5bbba0fb140f6b39c4f8906cd60e1023dded137d7d445ebdff8f03ecd9d","last_reissued_at":"2026-07-05T10:40:16.934329Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:16.934329Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging -- An Open Recipe","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kasima Tharnpipitchai, Kunat Pipatanakul, Pittawat Taveekitworachai, Potsawee Manakul","submitted_at":"2025-02-13T08:10:45Z","abstract_excerpt":"This paper investigates data selection and model merging methodologies aimed at incorporating advanced reasoning capabilities such as those of DeepSeek R1 into language-specific large language models (LLMs), with a particular focus on the Thai LLM. Our goal is to enhance the reasoning capabilities of language-specific LLMs while maintaining their target language abilities. DeepSeek R1 excels in reasoning but primarily benefits high-resource languages such as English and Chinese. However, low-resource languages remain underserved due to the dominance of English-centric training data and model o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09056","kind":"arxiv","version":3},"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/2502.09056/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":"2502.09056","created_at":"2026-07-05T10:40:16.934387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.09056v3","created_at":"2026-07-05T10:40:16.934387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09056","created_at":"2026-07-05T10:40:16.934387+00:00"},{"alias_kind":"pith_short_12","alias_value":"S376LO52B6YU","created_at":"2026-07-05T10:40:16.934387+00:00"},{"alias_kind":"pith_short_16","alias_value":"S376LO52B6YUB5VT","created_at":"2026-07-05T10:40:16.934387+00:00"},{"alias_kind":"pith_short_8","alias_value":"S376LO52","created_at":"2026-07-05T10:40:16.934387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":193,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02465","citing_title":"Learning When to Translate for Multilingual Reasoning","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD","json":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD.json","graph_json":"https://pith.science/api/pith-number/S376LO52B6YUB5VTTRHYSBWNMD/graph.json","events_json":"https://pith.science/api/pith-number/S376LO52B6YUB5VTTRHYSBWNMD/events.json","paper":"https://pith.science/paper/S376LO52"},"agent_actions":{"view_html":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD","download_json":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD.json","view_paper":"https://pith.science/paper/S376LO52","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.09056&json=true","fetch_graph":"https://pith.science/api/pith-number/S376LO52B6YUB5VTTRHYSBWNMD/graph.json","fetch_events":"https://pith.science/api/pith-number/S376LO52B6YUB5VTTRHYSBWNMD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD/action/storage_attestation","attest_author":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD/action/author_attestation","sign_citation":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD/action/citation_signature","submit_replication":"https://pith.science/pith/S376LO52B6YUB5VTTRHYSBWNMD/action/replication_record"}},"created_at":"2026-07-05T10:40:16.934387+00:00","updated_at":"2026-07-05T10:40:16.934387+00:00"}