{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HEK7RUC7X5BLZH33LLWE3VVVPN","short_pith_number":"pith:HEK7RUC7","schema_version":"1.0","canonical_sha256":"3915f8d05fbf42bc9f7b5aec4dd6b57b43c1e89f1b46d5e8ceb1099b24830344","source":{"kind":"arxiv","id":"2504.14810","version":2},"attestation_state":"computed","paper":{"title":"DONOD: Efficient and Generalizable Instruction Fine-Tuning for LLMs via Model-Intrinsic Dataset Pruning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dongzhan Zhou, Jucheng Hu, Lijun Wu, Surong Yang","submitted_at":"2025-04-21T02:25:03Z","abstract_excerpt":"Ad-hoc instruction fine-tuning of large language models (LLMs) is widely adopted for domain-specific adaptation. While domain-specific supervised fine-tuning (SFT) is effective and efficient, it often weakens cross-domain generalization and struggles with noisy training data. To address these challenges, we propose DONOD, a lightweight model-intrinsic data pruning method. Our approach evaluates data using two model-parameter-based metrics: Delta of Norm (DON), which captures the cumulative influence on model weights, and Norm of Delta (NOD), which quantifies weight instability. Moreover, by em"},"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":"2504.14810","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-21T02:25:03Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"29ed230329e04c12968858fa1596cc5cac6d87cbdebea89b1ef964707d34d577","abstract_canon_sha256":"7e66d505621d4aaade126d86521f81915a35e075b0c3ee00b87edf5d37b47ad8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:34.956519Z","signature_b64":"aMkIlsyh7chGQGTOd2ZOpBpsTWA5WCblvuFV5dphFcDBXk+gxyAIUBJhrd/eus5bzdL12j8DdbZ69TKI8YBxBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3915f8d05fbf42bc9f7b5aec4dd6b57b43c1e89f1b46d5e8ceb1099b24830344","last_reissued_at":"2026-07-05T11:50:34.956076Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:34.956076Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DONOD: Efficient and Generalizable Instruction Fine-Tuning for LLMs via Model-Intrinsic Dataset Pruning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dongzhan Zhou, Jucheng Hu, Lijun Wu, Surong Yang","submitted_at":"2025-04-21T02:25:03Z","abstract_excerpt":"Ad-hoc instruction fine-tuning of large language models (LLMs) is widely adopted for domain-specific adaptation. While domain-specific supervised fine-tuning (SFT) is effective and efficient, it often weakens cross-domain generalization and struggles with noisy training data. To address these challenges, we propose DONOD, a lightweight model-intrinsic data pruning method. Our approach evaluates data using two model-parameter-based metrics: Delta of Norm (DON), which captures the cumulative influence on model weights, and Norm of Delta (NOD), which quantifies weight instability. Moreover, by em"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.14810","kind":"arxiv","version":2},"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/2504.14810/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":"2504.14810","created_at":"2026-07-05T11:50:34.956134+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.14810v2","created_at":"2026-07-05T11:50:34.956134+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.14810","created_at":"2026-07-05T11:50:34.956134+00:00"},{"alias_kind":"pith_short_12","alias_value":"HEK7RUC7X5BL","created_at":"2026-07-05T11:50:34.956134+00:00"},{"alias_kind":"pith_short_16","alias_value":"HEK7RUC7X5BLZH33","created_at":"2026-07-05T11:50:34.956134+00:00"},{"alias_kind":"pith_short_8","alias_value":"HEK7RUC7","created_at":"2026-07-05T11:50:34.956134+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.12750","citing_title":"Multimodal-Guided Dynamic Dataset Pruning for Robust and Efficient Data-Centric Learning","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN","json":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN.json","graph_json":"https://pith.science/api/pith-number/HEK7RUC7X5BLZH33LLWE3VVVPN/graph.json","events_json":"https://pith.science/api/pith-number/HEK7RUC7X5BLZH33LLWE3VVVPN/events.json","paper":"https://pith.science/paper/HEK7RUC7"},"agent_actions":{"view_html":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN","download_json":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN.json","view_paper":"https://pith.science/paper/HEK7RUC7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.14810&json=true","fetch_graph":"https://pith.science/api/pith-number/HEK7RUC7X5BLZH33LLWE3VVVPN/graph.json","fetch_events":"https://pith.science/api/pith-number/HEK7RUC7X5BLZH33LLWE3VVVPN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN/action/storage_attestation","attest_author":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN/action/author_attestation","sign_citation":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN/action/citation_signature","submit_replication":"https://pith.science/pith/HEK7RUC7X5BLZH33LLWE3VVVPN/action/replication_record"}},"created_at":"2026-07-05T11:50:34.956134+00:00","updated_at":"2026-07-05T11:50:34.956134+00:00"}