{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MFKINZA35UO4ABNSJMBBNQQBFI","short_pith_number":"pith:MFKINZA3","schema_version":"1.0","canonical_sha256":"615486e41bed1dc005b24b0216c2012a3399cc5adf5b940b1cacaf0099eba9f9","source":{"kind":"arxiv","id":"2506.09446","version":1},"attestation_state":"computed","paper":{"title":"Harmonizing and Merging Source Models for CLIP-based Domain Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aihua Zheng, Bin Luo, Bo Jiang, Jian Liang, Yuhe Ding, Zi Wang","submitted_at":"2025-06-11T06:52:36Z","abstract_excerpt":"CLIP-based domain generalization aims to improve model generalization to unseen domains by leveraging the powerful zero-shot classification capabilities of CLIP and multiple source datasets. Existing methods typically train a single model across multiple source domains to capture domain-shared information. However, this paradigm inherently suffers from two types of conflicts: 1) sample conflicts, arising from noisy samples and extreme domain shifts among sources; and 2) optimization conflicts, stemming from competition and trade-offs during multi-source training. Both hinder the generalization"},"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":"2506.09446","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-11T06:52:36Z","cross_cats_sorted":[],"title_canon_sha256":"fe94996bc1448789efbc10fcc77ead0fd621157622458171bbbe6e553ffc5724","abstract_canon_sha256":"557672b95e3145bd3bd344b48139c5205f8490c573b321147d1762053f94c486"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:37.543363Z","signature_b64":"4YbyK41Az1P0G+3jz9xOFkIPU2nKWXTeKFeyOG1SHAHiBOHPcQZtiifnNgC9mkqR5q4D757YyLlq27ccnmRhCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"615486e41bed1dc005b24b0216c2012a3399cc5adf5b940b1cacaf0099eba9f9","last_reissued_at":"2026-07-05T11:19:37.542872Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:37.542872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Harmonizing and Merging Source Models for CLIP-based Domain Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aihua Zheng, Bin Luo, Bo Jiang, Jian Liang, Yuhe Ding, Zi Wang","submitted_at":"2025-06-11T06:52:36Z","abstract_excerpt":"CLIP-based domain generalization aims to improve model generalization to unseen domains by leveraging the powerful zero-shot classification capabilities of CLIP and multiple source datasets. Existing methods typically train a single model across multiple source domains to capture domain-shared information. However, this paradigm inherently suffers from two types of conflicts: 1) sample conflicts, arising from noisy samples and extreme domain shifts among sources; and 2) optimization conflicts, stemming from competition and trade-offs during multi-source training. Both hinder the generalization"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09446","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/2506.09446/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":"2506.09446","created_at":"2026-07-05T11:19:37.542929+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.09446v1","created_at":"2026-07-05T11:19:37.542929+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09446","created_at":"2026-07-05T11:19:37.542929+00:00"},{"alias_kind":"pith_short_12","alias_value":"MFKINZA35UO4","created_at":"2026-07-05T11:19:37.542929+00:00"},{"alias_kind":"pith_short_16","alias_value":"MFKINZA35UO4ABNS","created_at":"2026-07-05T11:19:37.542929+00:00"},{"alias_kind":"pith_short_8","alias_value":"MFKINZA3","created_at":"2026-07-05T11:19:37.542929+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30360","citing_title":"On the Vulnerability of Parameter-Level Defenses to Model Merging","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI","json":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI.json","graph_json":"https://pith.science/api/pith-number/MFKINZA35UO4ABNSJMBBNQQBFI/graph.json","events_json":"https://pith.science/api/pith-number/MFKINZA35UO4ABNSJMBBNQQBFI/events.json","paper":"https://pith.science/paper/MFKINZA3"},"agent_actions":{"view_html":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI","download_json":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI.json","view_paper":"https://pith.science/paper/MFKINZA3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.09446&json=true","fetch_graph":"https://pith.science/api/pith-number/MFKINZA35UO4ABNSJMBBNQQBFI/graph.json","fetch_events":"https://pith.science/api/pith-number/MFKINZA35UO4ABNSJMBBNQQBFI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI/action/storage_attestation","attest_author":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI/action/author_attestation","sign_citation":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI/action/citation_signature","submit_replication":"https://pith.science/pith/MFKINZA35UO4ABNSJMBBNQQBFI/action/replication_record"}},"created_at":"2026-07-05T11:19:37.542929+00:00","updated_at":"2026-07-05T11:19:37.542929+00:00"}