{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UYSBXLJSFK7ZNM6OW5HAHDFY2N","short_pith_number":"pith:UYSBXLJS","schema_version":"1.0","canonical_sha256":"a6241bad322abf96b3ceb74e038cb8d35b2054c9096e9965e03e119c81c88ba2","source":{"kind":"arxiv","id":"2405.04883","version":2},"attestation_state":"computed","paper":{"title":"FreeBind: Free Lunch in Unified Multimodal Space via Knowledge Fusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haifeng Huang, Luping Liu, Peng Gao, Rongjie Huang, Tao Jin, Xize Cheng, Yang Zhao, Zehan Wang, Zhenhui Ye, Zhou Zhao, Ziang Zhang","submitted_at":"2024-05-08T08:32:34Z","abstract_excerpt":"Unified multi-model representation spaces are the foundation of multimodal understanding and generation. However, the billions of model parameters and catastrophic forgetting problems make it challenging to further enhance pre-trained unified spaces. In this work, we propose FreeBind, an idea that treats multimodal representation spaces as basic units, and freely augments pre-trained unified space by integrating knowledge from extra expert spaces via \"space bonds\". Specifically, we introduce two kinds of basic space bonds: 1) Space Displacement Bond and 2) Space Combination Bond. Based on thes"},"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":"2405.04883","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-08T08:32:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3f609ed967b122eb8e90ca32ac0e56883534088a920ebbdeb428a04fbdc4a6d9","abstract_canon_sha256":"ad9355b0969b49960894269b8bb1db67e909668b76163ef7867ee2e793060226"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:17:33.218624Z","signature_b64":"W+ogukOxUxglTweU3bcJIvt7MpI6yNvNstQDChN/kVsB/9MkgodzI2WZWC/AyiCYjh3WmlkQ5aNt64hIgvgUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a6241bad322abf96b3ceb74e038cb8d35b2054c9096e9965e03e119c81c88ba2","last_reissued_at":"2026-07-05T08:17:33.218061Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:17:33.218061Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FreeBind: Free Lunch in Unified Multimodal Space via Knowledge Fusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haifeng Huang, Luping Liu, Peng Gao, Rongjie Huang, Tao Jin, Xize Cheng, Yang Zhao, Zehan Wang, Zhenhui Ye, Zhou Zhao, Ziang Zhang","submitted_at":"2024-05-08T08:32:34Z","abstract_excerpt":"Unified multi-model representation spaces are the foundation of multimodal understanding and generation. However, the billions of model parameters and catastrophic forgetting problems make it challenging to further enhance pre-trained unified spaces. In this work, we propose FreeBind, an idea that treats multimodal representation spaces as basic units, and freely augments pre-trained unified space by integrating knowledge from extra expert spaces via \"space bonds\". Specifically, we introduce two kinds of basic space bonds: 1) Space Displacement Bond and 2) Space Combination Bond. Based on thes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.04883","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/2405.04883/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":"2405.04883","created_at":"2026-07-05T08:17:33.218120+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.04883v2","created_at":"2026-07-05T08:17:33.218120+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.04883","created_at":"2026-07-05T08:17:33.218120+00:00"},{"alias_kind":"pith_short_12","alias_value":"UYSBXLJSFK7Z","created_at":"2026-07-05T08:17:33.218120+00:00"},{"alias_kind":"pith_short_16","alias_value":"UYSBXLJSFK7ZNM6O","created_at":"2026-07-05T08:17:33.218120+00:00"},{"alias_kind":"pith_short_8","alias_value":"UYSBXLJS","created_at":"2026-07-05T08:17:33.218120+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.12034","citing_title":"Calibrated Multimodal Representation Learning with Missing Modalities","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2511.21331","citing_title":"The More, the Merrier: Contrastive Fusion for Higher-Order Multimodal Alignment","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11043","citing_title":"EmergentBridge: Improving Zero-Shot Cross-Modal Transfer in Unified Multimodal Embedding Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11043","citing_title":"EmergentBridge: Improving Zero-Shot Cross-Modal Transfer in Unified Multimodal Embedding Models","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N","json":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N.json","graph_json":"https://pith.science/api/pith-number/UYSBXLJSFK7ZNM6OW5HAHDFY2N/graph.json","events_json":"https://pith.science/api/pith-number/UYSBXLJSFK7ZNM6OW5HAHDFY2N/events.json","paper":"https://pith.science/paper/UYSBXLJS"},"agent_actions":{"view_html":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N","download_json":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N.json","view_paper":"https://pith.science/paper/UYSBXLJS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.04883&json=true","fetch_graph":"https://pith.science/api/pith-number/UYSBXLJSFK7ZNM6OW5HAHDFY2N/graph.json","fetch_events":"https://pith.science/api/pith-number/UYSBXLJSFK7ZNM6OW5HAHDFY2N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N/action/storage_attestation","attest_author":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N/action/author_attestation","sign_citation":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N/action/citation_signature","submit_replication":"https://pith.science/pith/UYSBXLJSFK7ZNM6OW5HAHDFY2N/action/replication_record"}},"created_at":"2026-07-05T08:17:33.218120+00:00","updated_at":"2026-07-05T08:17:33.218120+00:00"}