{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZGVU5ZV3LWKIZANU6237R75KTH","short_pith_number":"pith:ZGVU5ZV3","schema_version":"1.0","canonical_sha256":"c9ab4ee6bb5d948c81b4f6b7f8ffaa99d3eb45968545666685ae28513d51e324","source":{"kind":"arxiv","id":"2408.09485","version":1},"attestation_state":"computed","paper":{"title":"Activated Parameter Locating via Causal Intervention for Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fanshuang Kong, Richong Zhang, Ziqiao Wang","submitted_at":"2024-08-18T14:00:00Z","abstract_excerpt":"Model merging combines multiple homologous models into one model, achieving convincing generalization without the necessity of additional training. A key challenge in this problem is resolving parameter redundancies and conflicts across multiple models. Existing models have demonstrated that dropping a portion of delta parameters can alleviate conflicts while maintaining performance. However, these methods often drop parameters either randomly or based on magnitude, overlooking task-specific information embedded in fine-tuned models. In this paper, we propose an Activated Parameter Locating (A"},"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":"2408.09485","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-18T14:00:00Z","cross_cats_sorted":[],"title_canon_sha256":"c876621e7eab6c38b6451f3df591e056e7fab34c90f8cb554439869de29f139c","abstract_canon_sha256":"f217dd2af8fbbad176deee31dc1cd78d4d9038934d50382fc106ed4179d6e3f5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:56:33.193799Z","signature_b64":"E/9fSY/4O2CAk+BYBx4KjFNPm2WP80yTSVXn5lHjzMbEHCwtnqTuJnIFkO+qG5ucWvmO866hW95EGm9ZLF7IAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9ab4ee6bb5d948c81b4f6b7f8ffaa99d3eb45968545666685ae28513d51e324","last_reissued_at":"2026-07-05T08:56:33.193375Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:56:33.193375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Activated Parameter Locating via Causal Intervention for Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fanshuang Kong, Richong Zhang, Ziqiao Wang","submitted_at":"2024-08-18T14:00:00Z","abstract_excerpt":"Model merging combines multiple homologous models into one model, achieving convincing generalization without the necessity of additional training. A key challenge in this problem is resolving parameter redundancies and conflicts across multiple models. Existing models have demonstrated that dropping a portion of delta parameters can alleviate conflicts while maintaining performance. However, these methods often drop parameters either randomly or based on magnitude, overlooking task-specific information embedded in fine-tuned models. In this paper, we propose an Activated Parameter Locating (A"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.09485","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/2408.09485/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":"2408.09485","created_at":"2026-07-05T08:56:33.193432+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.09485v1","created_at":"2026-07-05T08:56:33.193432+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.09485","created_at":"2026-07-05T08:56:33.193432+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZGVU5ZV3LWKI","created_at":"2026-07-05T08:56:33.193432+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZGVU5ZV3LWKIZANU","created_at":"2026-07-05T08:56:33.193432+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZGVU5ZV3","created_at":"2026-07-05T08:56:33.193432+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.07666","citing_title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","ref_index":112,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH","json":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH.json","graph_json":"https://pith.science/api/pith-number/ZGVU5ZV3LWKIZANU6237R75KTH/graph.json","events_json":"https://pith.science/api/pith-number/ZGVU5ZV3LWKIZANU6237R75KTH/events.json","paper":"https://pith.science/paper/ZGVU5ZV3"},"agent_actions":{"view_html":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH","download_json":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH.json","view_paper":"https://pith.science/paper/ZGVU5ZV3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.09485&json=true","fetch_graph":"https://pith.science/api/pith-number/ZGVU5ZV3LWKIZANU6237R75KTH/graph.json","fetch_events":"https://pith.science/api/pith-number/ZGVU5ZV3LWKIZANU6237R75KTH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH/action/storage_attestation","attest_author":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH/action/author_attestation","sign_citation":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH/action/citation_signature","submit_replication":"https://pith.science/pith/ZGVU5ZV3LWKIZANU6237R75KTH/action/replication_record"}},"created_at":"2026-07-05T08:56:33.193432+00:00","updated_at":"2026-07-05T08:56:33.193432+00:00"}