{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EDEZE32ZEZOCYFWHLDWJ6GAERQ","short_pith_number":"pith:EDEZE32Z","schema_version":"1.0","canonical_sha256":"20c9926f59265c2c16c758ec9f18048c02dd37bfdc1f07be9f3ba0da8589fa21","source":{"kind":"arxiv","id":"2411.04425","version":3},"attestation_state":"computed","paper":{"title":"DELIFT: Data Efficient Language model Instruction Fine Tuning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ishika Agarwal, Krishnateja Killamsetty, Lucian Popa, Marina Danilevksy","submitted_at":"2024-11-07T04:38:29Z","abstract_excerpt":"Fine-tuning large language models (LLMs) is essential for enhancing their performance on specific tasks but is often resource-intensive due to redundant or uninformative data. To address this inefficiency, we introduce DELIFT (Data Efficient Language model Instruction Fine-Tuning), a novel algorithm that systematically optimizes data selection across the three key stages of fine-tuning: (1) instruction tuning, (2) task-specific fine-tuning (e.g., reasoning, question-answering), and (3) continual fine-tuning (e.g., incorporating new data versions). Unlike existing methods that focus on single-s"},"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":"2411.04425","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T04:38:29Z","cross_cats_sorted":[],"title_canon_sha256":"7c31cce64d3fd2853361b1b6e9026527ae15cae401c5308df3749f79829f2a62","abstract_canon_sha256":"323ff22c65a4fb85e53c1b1cb929d6d0d28e3e4b046697e638d276163503c58d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:35:40.387795Z","signature_b64":"Eu2gVd2DzMNAHCdw+dkC6kVf63GV2yPV71NvjIVOItx2XjKYPa0c4tjWMdEqdJXZbEDQzdV51l6V8G3Dj0tEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20c9926f59265c2c16c758ec9f18048c02dd37bfdc1f07be9f3ba0da8589fa21","last_reissued_at":"2026-07-05T10:35:40.386830Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:35:40.386830Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DELIFT: Data Efficient Language model Instruction Fine Tuning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ishika Agarwal, Krishnateja Killamsetty, Lucian Popa, Marina Danilevksy","submitted_at":"2024-11-07T04:38:29Z","abstract_excerpt":"Fine-tuning large language models (LLMs) is essential for enhancing their performance on specific tasks but is often resource-intensive due to redundant or uninformative data. To address this inefficiency, we introduce DELIFT (Data Efficient Language model Instruction Fine-Tuning), a novel algorithm that systematically optimizes data selection across the three key stages of fine-tuning: (1) instruction tuning, (2) task-specific fine-tuning (e.g., reasoning, question-answering), and (3) continual fine-tuning (e.g., incorporating new data versions). Unlike existing methods that focus on single-s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04425","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/2411.04425/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":"2411.04425","created_at":"2026-07-05T10:35:40.386970+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.04425v3","created_at":"2026-07-05T10:35:40.386970+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04425","created_at":"2026-07-05T10:35:40.386970+00:00"},{"alias_kind":"pith_short_12","alias_value":"EDEZE32ZEZOC","created_at":"2026-07-05T10:35:40.386970+00:00"},{"alias_kind":"pith_short_16","alias_value":"EDEZE32ZEZOCYFWH","created_at":"2026-07-05T10:35:40.386970+00:00"},{"alias_kind":"pith_short_8","alias_value":"EDEZE32Z","created_at":"2026-07-05T10:35:40.386970+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.02153","citing_title":"Small Language Models are the Future of Agentic AI","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04428","citing_title":"Submodular Ground-Set Pruning: Monotone Tightness and a Non-Monotone Separation","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ","json":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ.json","graph_json":"https://pith.science/api/pith-number/EDEZE32ZEZOCYFWHLDWJ6GAERQ/graph.json","events_json":"https://pith.science/api/pith-number/EDEZE32ZEZOCYFWHLDWJ6GAERQ/events.json","paper":"https://pith.science/paper/EDEZE32Z"},"agent_actions":{"view_html":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ","download_json":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ.json","view_paper":"https://pith.science/paper/EDEZE32Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.04425&json=true","fetch_graph":"https://pith.science/api/pith-number/EDEZE32ZEZOCYFWHLDWJ6GAERQ/graph.json","fetch_events":"https://pith.science/api/pith-number/EDEZE32ZEZOCYFWHLDWJ6GAERQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ/action/storage_attestation","attest_author":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ/action/author_attestation","sign_citation":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ/action/citation_signature","submit_replication":"https://pith.science/pith/EDEZE32ZEZOCYFWHLDWJ6GAERQ/action/replication_record"}},"created_at":"2026-07-05T10:35:40.386970+00:00","updated_at":"2026-07-05T10:35:40.386970+00:00"}