{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CD7QEITGN2GAUSVMD2VXNJHMPW","short_pith_number":"pith:CD7QEITG","schema_version":"1.0","canonical_sha256":"10ff0222666e8c0a4aac1eab76a4ec7d9c7c97ef5898cadeed09743d18f4752b","source":{"kind":"arxiv","id":"2502.12217","version":1},"attestation_state":"computed","paper":{"title":"Optimal Brain Iterative Merging: Mitigating Interference in LLM Merging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Biye Li, Yixuan Qiao, Yunfang Wu, Zhenyu Mao, Zhixiang Wang","submitted_at":"2025-02-17T09:07:49Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities, but their high computational costs pose challenges for customization. Model merging offers a cost-effective alternative, yet existing methods suffer from interference among parameters, leading to performance degradation. In this work, we propose Optimal Brain Iterative Merging (OBIM), a novel method designed to mitigate both intra-model and inter-model interference. OBIM consists of two key components: (1) A saliency measurement mechanism that evaluates parameter importance based on loss changes induced by individual weigh"},"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":"2502.12217","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-17T09:07:49Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ff1b619bff0d01516db7cd3d957e3df84d9ae47a2f305eb5cec514cc543582c5","abstract_canon_sha256":"199a04126804e4742adb74acbdd988d819695114a02c49956643a030c415b361"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:15:52.166253Z","signature_b64":"US0kMAf8KNrHAwyo0/1Ppm1s0wgRwWbLcJbGUNVBvrps+8blakFTmReLQ3OwqD+4+hKn/DhluvyPFxygw4q1BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10ff0222666e8c0a4aac1eab76a4ec7d9c7c97ef5898cadeed09743d18f4752b","last_reissued_at":"2026-07-05T10:15:52.165755Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:15:52.165755Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimal Brain Iterative Merging: Mitigating Interference in LLM Merging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Biye Li, Yixuan Qiao, Yunfang Wu, Zhenyu Mao, Zhixiang Wang","submitted_at":"2025-02-17T09:07:49Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities, but their high computational costs pose challenges for customization. Model merging offers a cost-effective alternative, yet existing methods suffer from interference among parameters, leading to performance degradation. In this work, we propose Optimal Brain Iterative Merging (OBIM), a novel method designed to mitigate both intra-model and inter-model interference. OBIM consists of two key components: (1) A saliency measurement mechanism that evaluates parameter importance based on loss changes induced by individual weigh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12217","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/2502.12217/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":"2502.12217","created_at":"2026-07-05T10:15:52.165807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.12217v1","created_at":"2026-07-05T10:15:52.165807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12217","created_at":"2026-07-05T10:15:52.165807+00:00"},{"alias_kind":"pith_short_12","alias_value":"CD7QEITGN2GA","created_at":"2026-07-05T10:15:52.165807+00:00"},{"alias_kind":"pith_short_16","alias_value":"CD7QEITGN2GAUSVM","created_at":"2026-07-05T10:15:52.165807+00:00"},{"alias_kind":"pith_short_8","alias_value":"CD7QEITG","created_at":"2026-07-05T10:15:52.165807+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW","json":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW.json","graph_json":"https://pith.science/api/pith-number/CD7QEITGN2GAUSVMD2VXNJHMPW/graph.json","events_json":"https://pith.science/api/pith-number/CD7QEITGN2GAUSVMD2VXNJHMPW/events.json","paper":"https://pith.science/paper/CD7QEITG"},"agent_actions":{"view_html":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW","download_json":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW.json","view_paper":"https://pith.science/paper/CD7QEITG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.12217&json=true","fetch_graph":"https://pith.science/api/pith-number/CD7QEITGN2GAUSVMD2VXNJHMPW/graph.json","fetch_events":"https://pith.science/api/pith-number/CD7QEITGN2GAUSVMD2VXNJHMPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW/action/storage_attestation","attest_author":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW/action/author_attestation","sign_citation":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW/action/citation_signature","submit_replication":"https://pith.science/pith/CD7QEITGN2GAUSVMD2VXNJHMPW/action/replication_record"}},"created_at":"2026-07-05T10:15:52.165807+00:00","updated_at":"2026-07-05T10:15:52.165807+00:00"}