{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NJPUXRYRF65RGE236L6G7YLY2G","short_pith_number":"pith:NJPUXRYR","schema_version":"1.0","canonical_sha256":"6a5f4bc7112fbb13135bf2fc6fe178d1bf9590398d188a6fc6485743dfbfe540","source":{"kind":"arxiv","id":"2407.13906","version":1},"attestation_state":"computed","paper":{"title":"Crafting Efficient Fine-Tuning Strategies for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Guan Wang, Michael Oliver","submitted_at":"2024-07-18T21:36:00Z","abstract_excerpt":"This paper addresses the challenges of efficiently fine-tuning large language models (LLMs) by exploring data efficiency and hyperparameter optimization. We investigate the minimum data required for effective fine-tuning and propose a novel hyperparameter optimization method that leverages early-stage model performance. Our experiments demonstrate that fine-tuning with as few as 200 samples can improve model accuracy from 70\\% to 88\\% in a product attribute extraction task. We identify a saturation point of approximately 6,500 samples, beyond which additional data yields diminishing returns. 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":"2407.13906","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-18T21:36:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"902076e8b79ba57fd8457588baa84cd3bd64699d47b883426e9d426057b2bbb1","abstract_canon_sha256":"b2b091433ca5315bdf940b7c025d8d988301582400259a0d93748f6367547baa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:45:55.313339Z","signature_b64":"dTv3JWkEmOTKyfTLNon4+KOSB0TgcPGb8YfMc5thYX59+r3yyv1ksj48bKU35OD3aPcxDqr7lerOOb9SwuFaCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6a5f4bc7112fbb13135bf2fc6fe178d1bf9590398d188a6fc6485743dfbfe540","last_reissued_at":"2026-07-05T08:45:55.312841Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:45:55.312841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Crafting Efficient Fine-Tuning Strategies for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Guan Wang, Michael Oliver","submitted_at":"2024-07-18T21:36:00Z","abstract_excerpt":"This paper addresses the challenges of efficiently fine-tuning large language models (LLMs) by exploring data efficiency and hyperparameter optimization. We investigate the minimum data required for effective fine-tuning and propose a novel hyperparameter optimization method that leverages early-stage model performance. Our experiments demonstrate that fine-tuning with as few as 200 samples can improve model accuracy from 70\\% to 88\\% in a product attribute extraction task. We identify a saturation point of approximately 6,500 samples, beyond which additional data yields diminishing returns. O"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.13906","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/2407.13906/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":"2407.13906","created_at":"2026-07-05T08:45:55.312905+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.13906v1","created_at":"2026-07-05T08:45:55.312905+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.13906","created_at":"2026-07-05T08:45:55.312905+00:00"},{"alias_kind":"pith_short_12","alias_value":"NJPUXRYRF65R","created_at":"2026-07-05T08:45:55.312905+00:00"},{"alias_kind":"pith_short_16","alias_value":"NJPUXRYRF65RGE23","created_at":"2026-07-05T08:45:55.312905+00:00"},{"alias_kind":"pith_short_8","alias_value":"NJPUXRYR","created_at":"2026-07-05T08:45:55.312905+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.23267","citing_title":"Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23267","citing_title":"Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G","json":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G.json","graph_json":"https://pith.science/api/pith-number/NJPUXRYRF65RGE236L6G7YLY2G/graph.json","events_json":"https://pith.science/api/pith-number/NJPUXRYRF65RGE236L6G7YLY2G/events.json","paper":"https://pith.science/paper/NJPUXRYR"},"agent_actions":{"view_html":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G","download_json":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G.json","view_paper":"https://pith.science/paper/NJPUXRYR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.13906&json=true","fetch_graph":"https://pith.science/api/pith-number/NJPUXRYRF65RGE236L6G7YLY2G/graph.json","fetch_events":"https://pith.science/api/pith-number/NJPUXRYRF65RGE236L6G7YLY2G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G/action/storage_attestation","attest_author":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G/action/author_attestation","sign_citation":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G/action/citation_signature","submit_replication":"https://pith.science/pith/NJPUXRYRF65RGE236L6G7YLY2G/action/replication_record"}},"created_at":"2026-07-05T08:45:55.312905+00:00","updated_at":"2026-07-05T08:45:55.312905+00:00"}