{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2AZZERRD2CUXUJVQKHAA5AUTL3","short_pith_number":"pith:2AZZERRD","schema_version":"1.0","canonical_sha256":"d033924623d0a97a26b051c00e82935ec21af57d3b8b928f083abd468a151e41","source":{"kind":"arxiv","id":"2204.02485","version":1},"attestation_state":"computed","paper":{"title":"Training-Free Robust Multimodal Learning via Sample-Wise Jacobian Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Hang Zhao, Siting Li, Sucheng Ren, Zhengqi Gao, Zihui Xue","submitted_at":"2022-04-05T20:38:33Z","abstract_excerpt":"Multimodal fusion emerges as an appealing technique to improve model performances on many tasks. Nevertheless, the robustness of such fusion methods is rarely involved in the present literature. In this paper, we propose a training-free robust late-fusion method by exploiting conditional independence assumption and Jacobian regularization. Our key is to minimize the Frobenius norm of a Jacobian matrix, where the resulting optimization problem is relaxed to a tractable Sylvester equation. Furthermore, we provide a theoretical error bound of our method and some insights about the function of the"},"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":"2204.02485","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-04-05T20:38:33Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"cb6dacef107d891eeb4e015f3abc31dda299621ae8fc8fe674925734e0bfa912","abstract_canon_sha256":"c66b02bfd098ea32f8e31e03abc16a064c498e424a97cd7b288abf694fd07fbb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:08.816549Z","signature_b64":"RLG9kNzOTvNzATTqRiyZxpUGJ3LtQyAe99tm3qQ43Tn9JG3Ml4jZhzrHvSbceCF0FxBC4XtikVyl5mVfkNFyCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d033924623d0a97a26b051c00e82935ec21af57d3b8b928f083abd468a151e41","last_reissued_at":"2026-07-05T04:12:08.816168Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:08.816168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training-Free Robust Multimodal Learning via Sample-Wise Jacobian Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Hang Zhao, Siting Li, Sucheng Ren, Zhengqi Gao, Zihui Xue","submitted_at":"2022-04-05T20:38:33Z","abstract_excerpt":"Multimodal fusion emerges as an appealing technique to improve model performances on many tasks. Nevertheless, the robustness of such fusion methods is rarely involved in the present literature. In this paper, we propose a training-free robust late-fusion method by exploiting conditional independence assumption and Jacobian regularization. Our key is to minimize the Frobenius norm of a Jacobian matrix, where the resulting optimization problem is relaxed to a tractable Sylvester equation. Furthermore, we provide a theoretical error bound of our method and some insights about the function of the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.02485","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/2204.02485/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":"2204.02485","created_at":"2026-07-05T04:12:08.816227+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.02485v1","created_at":"2026-07-05T04:12:08.816227+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.02485","created_at":"2026-07-05T04:12:08.816227+00:00"},{"alias_kind":"pith_short_12","alias_value":"2AZZERRD2CUX","created_at":"2026-07-05T04:12:08.816227+00:00"},{"alias_kind":"pith_short_16","alias_value":"2AZZERRD2CUXUJVQ","created_at":"2026-07-05T04:12:08.816227+00:00"},{"alias_kind":"pith_short_8","alias_value":"2AZZERRD","created_at":"2026-07-05T04:12:08.816227+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/2AZZERRD2CUXUJVQKHAA5AUTL3","json":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3.json","graph_json":"https://pith.science/api/pith-number/2AZZERRD2CUXUJVQKHAA5AUTL3/graph.json","events_json":"https://pith.science/api/pith-number/2AZZERRD2CUXUJVQKHAA5AUTL3/events.json","paper":"https://pith.science/paper/2AZZERRD"},"agent_actions":{"view_html":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3","download_json":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3.json","view_paper":"https://pith.science/paper/2AZZERRD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.02485&json=true","fetch_graph":"https://pith.science/api/pith-number/2AZZERRD2CUXUJVQKHAA5AUTL3/graph.json","fetch_events":"https://pith.science/api/pith-number/2AZZERRD2CUXUJVQKHAA5AUTL3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3/action/storage_attestation","attest_author":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3/action/author_attestation","sign_citation":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3/action/citation_signature","submit_replication":"https://pith.science/pith/2AZZERRD2CUXUJVQKHAA5AUTL3/action/replication_record"}},"created_at":"2026-07-05T04:12:08.816227+00:00","updated_at":"2026-07-05T04:12:08.816227+00:00"}