{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3UIWKHMNIERBF7AMO67IZURIZO","short_pith_number":"pith:3UIWKHMN","schema_version":"1.0","canonical_sha256":"dd11651d8d412212fc0c77be8cd228cba5e53fe04514810a95efdc80ee949b31","source":{"kind":"arxiv","id":"2412.13337","version":1},"attestation_state":"computed","paper":{"title":"Unveiling the Secret Recipe: A Guide For Supervised Fine-Tuning Small LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhishek Bhandwaldar, Akash Srivastava, Aldo Pareja, Guangxuan Xu, Hao Wang, Kai Xu, Krishnateja Killamsetty, Ligong Han, Luke Inglis, Nikhil Shivakumar Nayak, Seungwook Han, Shivchander Sudalairaj, Wenlong Zhao","submitted_at":"2024-12-17T21:16:59Z","abstract_excerpt":"The rise of large language models (LLMs) has created a significant disparity: industrial research labs with their computational resources, expert teams, and advanced infrastructures, can effectively fine-tune LLMs, while individual developers and small organizations face barriers due to limited resources. In this paper, we aim to bridge this gap by presenting a comprehensive study on supervised fine-tuning of LLMs using instruction-tuning datasets spanning diverse knowledge domains and skills. We focus on small-sized LLMs (3B to 7B parameters) for their cost-efficiency and accessibility. We ex"},"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":"2412.13337","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-17T21:16:59Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"47dbc308e5cceaf389c3ac59f7a188c2a051d9e2d75bf9569586b208857dc230","abstract_canon_sha256":"56cd686ee5e467512f2b61d0d5b01fb9fd9c88425da1dd125346cd33914f34b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:50:59.894593Z","signature_b64":"QBX+OQMhyW0KiZymr/Jb58+O/lOpXQedgNz3M1IQCIpvPxg5897lS5e61Ps5500u8FIabVbTYnmtm9MKZ/AmBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dd11651d8d412212fc0c77be8cd228cba5e53fe04514810a95efdc80ee949b31","last_reissued_at":"2026-07-05T09:50:59.894136Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:50:59.894136Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unveiling the Secret Recipe: A Guide For Supervised Fine-Tuning Small LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhishek Bhandwaldar, Akash Srivastava, Aldo Pareja, Guangxuan Xu, Hao Wang, Kai Xu, Krishnateja Killamsetty, Ligong Han, Luke Inglis, Nikhil Shivakumar Nayak, Seungwook Han, Shivchander Sudalairaj, Wenlong Zhao","submitted_at":"2024-12-17T21:16:59Z","abstract_excerpt":"The rise of large language models (LLMs) has created a significant disparity: industrial research labs with their computational resources, expert teams, and advanced infrastructures, can effectively fine-tune LLMs, while individual developers and small organizations face barriers due to limited resources. In this paper, we aim to bridge this gap by presenting a comprehensive study on supervised fine-tuning of LLMs using instruction-tuning datasets spanning diverse knowledge domains and skills. We focus on small-sized LLMs (3B to 7B parameters) for their cost-efficiency and accessibility. We ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13337","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/2412.13337/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":"2412.13337","created_at":"2026-07-05T09:50:59.894208+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.13337v1","created_at":"2026-07-05T09:50:59.894208+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13337","created_at":"2026-07-05T09:50:59.894208+00:00"},{"alias_kind":"pith_short_12","alias_value":"3UIWKHMNIERB","created_at":"2026-07-05T09:50:59.894208+00:00"},{"alias_kind":"pith_short_16","alias_value":"3UIWKHMNIERBF7AM","created_at":"2026-07-05T09:50:59.894208+00:00"},{"alias_kind":"pith_short_8","alias_value":"3UIWKHMN","created_at":"2026-07-05T09:50:59.894208+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.07177","citing_title":"Towards EnergyGPT: A Large Language Model Specialized for the Energy Sector","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2509.13047","citing_title":"Multi-Model Synthetic Training for Mission-Critical Small Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2512.03053","citing_title":"Mitigating hallucinations and omissions in LLMs for invertible problems: An application to hardware logic design automation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2601.22264","citing_title":"Predicting Intermittent Job Failure Categories for Diagnosis Using Few-Shot Fine-Tuned Language Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09791","citing_title":"Pioneer Agent: Continual Improvement of Small Language Models in Production","ref_index":71,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO","json":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO.json","graph_json":"https://pith.science/api/pith-number/3UIWKHMNIERBF7AMO67IZURIZO/graph.json","events_json":"https://pith.science/api/pith-number/3UIWKHMNIERBF7AMO67IZURIZO/events.json","paper":"https://pith.science/paper/3UIWKHMN"},"agent_actions":{"view_html":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO","download_json":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO.json","view_paper":"https://pith.science/paper/3UIWKHMN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.13337&json=true","fetch_graph":"https://pith.science/api/pith-number/3UIWKHMNIERBF7AMO67IZURIZO/graph.json","fetch_events":"https://pith.science/api/pith-number/3UIWKHMNIERBF7AMO67IZURIZO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO/action/storage_attestation","attest_author":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO/action/author_attestation","sign_citation":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO/action/citation_signature","submit_replication":"https://pith.science/pith/3UIWKHMNIERBF7AMO67IZURIZO/action/replication_record"}},"created_at":"2026-07-05T09:50:59.894208+00:00","updated_at":"2026-07-05T09:50:59.894208+00:00"}