{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GXU3XXDEJSUM5ZGE6H7DZK4HMF","short_pith_number":"pith:GXU3XXDE","schema_version":"1.0","canonical_sha256":"35e9bbdc644ca8cee4c4f1fe3cab8761508e2fe6ca09fded3b3c46c7d9e3bb31","source":{"kind":"arxiv","id":"2606.09331","version":1},"attestation_state":"computed","paper":{"title":"Conan-embedding-v3: Fusing Modality-Specific Models for Omni-Modal Embedding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.MM","authors_text":"Peiming Li, Shiyu Li, Yang Tang, Yifan Wang, Zheng Wei, Zhiyuan Hu","submitted_at":"2026-06-08T10:54:18Z","abstract_excerpt":"Omni-modal retrieval promises a single embedding space for text, image, video, document, and audio inputs, but building such a unified retriever is difficult since these modalities differ in data distribution, architecture, and optimization dynamics. In this work, we present Conan-embedding-v3, a decouple--fuse--recover framework for omni-modal retrieval. Conan-embedding-v3 first trains modality specialists independently and fuses their task vectors into a single dense backbone, a strategy we call Decoupled Specialist Fusion. We show that this fusion composes visual, video, and document retrie"},"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":"2606.09331","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.MM","submitted_at":"2026-06-08T10:54:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2bf26518ee159bc1506f22ec203784ed67b1955420b8df83a111ccba14c782fc","abstract_canon_sha256":"a736a9be6fe0ec3a137fd1b4256b1bd3cc051ca53519b2ed4fa1c66de1bd8c79"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:08:15.624503Z","signature_b64":"RvIEHXAA7qPe/ZQBRciZTeBteOj0ZGOWxHlgb9bty9DRALuH0JQBBMZIlr4jR7SniDidwJvL/2AHT6GO+9xoCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35e9bbdc644ca8cee4c4f1fe3cab8761508e2fe6ca09fded3b3c46c7d9e3bb31","last_reissued_at":"2026-06-09T02:08:15.623058Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:08:15.623058Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conan-embedding-v3: Fusing Modality-Specific Models for Omni-Modal Embedding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.MM","authors_text":"Peiming Li, Shiyu Li, Yang Tang, Yifan Wang, Zheng Wei, Zhiyuan Hu","submitted_at":"2026-06-08T10:54:18Z","abstract_excerpt":"Omni-modal retrieval promises a single embedding space for text, image, video, document, and audio inputs, but building such a unified retriever is difficult since these modalities differ in data distribution, architecture, and optimization dynamics. In this work, we present Conan-embedding-v3, a decouple--fuse--recover framework for omni-modal retrieval. Conan-embedding-v3 first trains modality specialists independently and fuses their task vectors into a single dense backbone, a strategy we call Decoupled Specialist Fusion. We show that this fusion composes visual, video, and document retrie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09331","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/2606.09331/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":"2606.09331","created_at":"2026-06-09T02:08:15.623218+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09331v1","created_at":"2026-06-09T02:08:15.623218+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09331","created_at":"2026-06-09T02:08:15.623218+00:00"},{"alias_kind":"pith_short_12","alias_value":"GXU3XXDEJSUM","created_at":"2026-06-09T02:08:15.623218+00:00"},{"alias_kind":"pith_short_16","alias_value":"GXU3XXDEJSUM5ZGE","created_at":"2026-06-09T02:08:15.623218+00:00"},{"alias_kind":"pith_short_8","alias_value":"GXU3XXDE","created_at":"2026-06-09T02:08:15.623218+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/GXU3XXDEJSUM5ZGE6H7DZK4HMF","json":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF.json","graph_json":"https://pith.science/api/pith-number/GXU3XXDEJSUM5ZGE6H7DZK4HMF/graph.json","events_json":"https://pith.science/api/pith-number/GXU3XXDEJSUM5ZGE6H7DZK4HMF/events.json","paper":"https://pith.science/paper/GXU3XXDE"},"agent_actions":{"view_html":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF","download_json":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF.json","view_paper":"https://pith.science/paper/GXU3XXDE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09331&json=true","fetch_graph":"https://pith.science/api/pith-number/GXU3XXDEJSUM5ZGE6H7DZK4HMF/graph.json","fetch_events":"https://pith.science/api/pith-number/GXU3XXDEJSUM5ZGE6H7DZK4HMF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF/action/storage_attestation","attest_author":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF/action/author_attestation","sign_citation":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF/action/citation_signature","submit_replication":"https://pith.science/pith/GXU3XXDEJSUM5ZGE6H7DZK4HMF/action/replication_record"}},"created_at":"2026-06-09T02:08:15.623218+00:00","updated_at":"2026-06-09T02:08:15.623218+00:00"}