{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ER7P2AL6ZHRUJDSOEPAVITJVFM","short_pith_number":"pith:ER7P2AL6","schema_version":"1.0","canonical_sha256":"247efd017ec9e3448e4e23c1544d352b121a42e10070858637a41c24d35f4dfd","source":{"kind":"arxiv","id":"2104.02000","version":1},"attestation_state":"computed","paper":{"title":"Can audio-visual integration strengthen robustness under multimodal attacks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Chenliang Xu, Yapeng Tian","submitted_at":"2021-04-05T16:46:45Z","abstract_excerpt":"In this paper, we propose to make a systematic study on machines multisensory perception under attacks. We use the audio-visual event recognition task against multimodal adversarial attacks as a proxy to investigate the robustness of audio-visual learning. We attack audio, visual, and both modalities to explore whether audio-visual integration still strengthens perception and how different fusion mechanisms affect the robustness of audio-visual models. For interpreting the multimodal interactions under attacks, we learn a weakly-supervised sound source visual localization model to localize sou"},"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":"2104.02000","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-04-05T16:46:45Z","cross_cats_sorted":["cs.CR","cs.SD","eess.AS"],"title_canon_sha256":"e424d42760c4a0f01ed5a0f27322488938934d0dc6704e79cb13ea05dde03637","abstract_canon_sha256":"c2cee036778baefd4f2527b2a5f6f4af375d3cbe70a5b846d690e0bba81520b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:29:06.314989Z","signature_b64":"Gi326i2TPBnV9Vqz4bQnrLiGE2XGOxjR9RdlMnbWNjwklemleSy2EqMmsHKOUIe2TQ5sJq9tPt+SdUlz1uuXBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"247efd017ec9e3448e4e23c1544d352b121a42e10070858637a41c24d35f4dfd","last_reissued_at":"2026-07-05T02:29:06.314512Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:29:06.314512Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can audio-visual integration strengthen robustness under multimodal attacks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Chenliang Xu, Yapeng Tian","submitted_at":"2021-04-05T16:46:45Z","abstract_excerpt":"In this paper, we propose to make a systematic study on machines multisensory perception under attacks. We use the audio-visual event recognition task against multimodal adversarial attacks as a proxy to investigate the robustness of audio-visual learning. We attack audio, visual, and both modalities to explore whether audio-visual integration still strengthens perception and how different fusion mechanisms affect the robustness of audio-visual models. For interpreting the multimodal interactions under attacks, we learn a weakly-supervised sound source visual localization model to localize sou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.02000","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/2104.02000/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":"2104.02000","created_at":"2026-07-05T02:29:06.314571+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.02000v1","created_at":"2026-07-05T02:29:06.314571+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.02000","created_at":"2026-07-05T02:29:06.314571+00:00"},{"alias_kind":"pith_short_12","alias_value":"ER7P2AL6ZHRU","created_at":"2026-07-05T02:29:06.314571+00:00"},{"alias_kind":"pith_short_16","alias_value":"ER7P2AL6ZHRUJDSO","created_at":"2026-07-05T02:29:06.314571+00:00"},{"alias_kind":"pith_short_8","alias_value":"ER7P2AL6","created_at":"2026-07-05T02:29:06.314571+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/ER7P2AL6ZHRUJDSOEPAVITJVFM","json":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM.json","graph_json":"https://pith.science/api/pith-number/ER7P2AL6ZHRUJDSOEPAVITJVFM/graph.json","events_json":"https://pith.science/api/pith-number/ER7P2AL6ZHRUJDSOEPAVITJVFM/events.json","paper":"https://pith.science/paper/ER7P2AL6"},"agent_actions":{"view_html":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM","download_json":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM.json","view_paper":"https://pith.science/paper/ER7P2AL6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.02000&json=true","fetch_graph":"https://pith.science/api/pith-number/ER7P2AL6ZHRUJDSOEPAVITJVFM/graph.json","fetch_events":"https://pith.science/api/pith-number/ER7P2AL6ZHRUJDSOEPAVITJVFM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM/action/storage_attestation","attest_author":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM/action/author_attestation","sign_citation":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM/action/citation_signature","submit_replication":"https://pith.science/pith/ER7P2AL6ZHRUJDSOEPAVITJVFM/action/replication_record"}},"created_at":"2026-07-05T02:29:06.314571+00:00","updated_at":"2026-07-05T02:29:06.314571+00:00"}