{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:T3XC6ISQEEMLF6JA6NKJWRBYIM","short_pith_number":"pith:T3XC6ISQ","schema_version":"1.0","canonical_sha256":"9eee2f22502118b2f920f3549b4438432f06d41b6079af00c766be66cb6a4f83","source":{"kind":"arxiv","id":"2306.02050","version":2},"attestation_state":"computed","paper":{"title":"Provable Dynamic Fusion for Low-Quality Multimodal Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Changqing Zhang, Haitao Wu, Huazhu Fu, Joey Tianyi Zhou, Qinghua Hu, Qingyang Zhang, Xi Peng","submitted_at":"2023-06-03T08:32:35Z","abstract_excerpt":"The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the gene"},"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":"2306.02050","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-03T08:32:35Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"00001f6c11effc02b12ab86567d0e121cb8c97fe5984d2b3d950b42e975bbad0","abstract_canon_sha256":"e7698701a6add0a073da0b3b452ca30e984e76192238c602e9bdd06b3da39368"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:17:53.071461Z","signature_b64":"U4sC3c+g4n1N3889W2a4nzZAyQMeuNTADV4mqIlG8w3QviBALz2i2gXGjMr+i6iGi3pY7za1VyXgdlI4Sj/rBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9eee2f22502118b2f920f3549b4438432f06d41b6079af00c766be66cb6a4f83","last_reissued_at":"2026-07-05T06:17:53.070981Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:17:53.070981Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Provable Dynamic Fusion for Low-Quality Multimodal Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Changqing Zhang, Haitao Wu, Huazhu Fu, Joey Tianyi Zhou, Qinghua Hu, Qingyang Zhang, Xi Peng","submitted_at":"2023-06-03T08:32:35Z","abstract_excerpt":"The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the gene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.02050","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/2306.02050/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":"2306.02050","created_at":"2026-07-05T06:17:53.071039+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.02050v2","created_at":"2026-07-05T06:17:53.071039+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.02050","created_at":"2026-07-05T06:17:53.071039+00:00"},{"alias_kind":"pith_short_12","alias_value":"T3XC6ISQEEML","created_at":"2026-07-05T06:17:53.071039+00:00"},{"alias_kind":"pith_short_16","alias_value":"T3XC6ISQEEMLF6JA","created_at":"2026-07-05T06:17:53.071039+00:00"},{"alias_kind":"pith_short_8","alias_value":"T3XC6ISQ","created_at":"2026-07-05T06:17:53.071039+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.24602","citing_title":"Majorization-Guided Test-Time Adaptation for Vision-Language Models under Modality-Specific Shift","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM","json":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM.json","graph_json":"https://pith.science/api/pith-number/T3XC6ISQEEMLF6JA6NKJWRBYIM/graph.json","events_json":"https://pith.science/api/pith-number/T3XC6ISQEEMLF6JA6NKJWRBYIM/events.json","paper":"https://pith.science/paper/T3XC6ISQ"},"agent_actions":{"view_html":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM","download_json":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM.json","view_paper":"https://pith.science/paper/T3XC6ISQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.02050&json=true","fetch_graph":"https://pith.science/api/pith-number/T3XC6ISQEEMLF6JA6NKJWRBYIM/graph.json","fetch_events":"https://pith.science/api/pith-number/T3XC6ISQEEMLF6JA6NKJWRBYIM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM/action/storage_attestation","attest_author":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM/action/author_attestation","sign_citation":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM/action/citation_signature","submit_replication":"https://pith.science/pith/T3XC6ISQEEMLF6JA6NKJWRBYIM/action/replication_record"}},"created_at":"2026-07-05T06:17:53.071039+00:00","updated_at":"2026-07-05T06:17:53.071039+00:00"}