{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:R26JHEFS7GHKYAP4BCUFOAGXRF","short_pith_number":"pith:R26JHEFS","schema_version":"1.0","canonical_sha256":"8ebc9390b2f98eac01fc08a85700d78975fb4afd99a0327ec3e66067550312a1","source":{"kind":"arxiv","id":"2507.11968","version":1},"attestation_state":"computed","paper":{"title":"Watch, Listen, Understand, Mislead: Tri-modal Adversarial Attacks on Short Videos for Content Appropriateness Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mohammed Jawwadul Islam, Montasir Chowdhury, Sadia Ahmmed, Sahid Hossain Mustakim, S M Jishanul Islam, Swakkhar Shatabda, Syed Tasdid Azam Dhrubo, Tashfia Sikder, Ummay Maria Muna","submitted_at":"2025-07-16T07:02:15Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) are increasingly used for content moderation, yet their robustness in short-form video contexts remains underexplored. Current safety evaluations often rely on unimodal attacks, failing to address combined attack vulnerabilities. In this paper, we introduce a comprehensive framework for evaluating the tri-modal safety of MLLMs. First, we present the Short-Video Multimodal Adversarial (SVMA) dataset, comprising diverse short-form videos with human-guided synthetic adversarial attacks. Second, we propose ChimeraBreak, a novel tri-modal attack strategy tha"},"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":"2507.11968","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-16T07:02:15Z","cross_cats_sorted":[],"title_canon_sha256":"e55d63874b57604cab82373273249c10cbcb1afa27d73a77a47ff47b4b929d4a","abstract_canon_sha256":"4691ff36167969eef341f51d1bdd890b29603ce8189c16a86ed5c299c25446de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:12.025696Z","signature_b64":"eeJEjNEJW3CrXZcRk+Mc5RJpvWmMuvCpn0T8q5RDYo1Mj9EhUCxsoAk+zBlKzuMQP+tsFX+y6JYACxyyCbAEAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ebc9390b2f98eac01fc08a85700d78975fb4afd99a0327ec3e66067550312a1","last_reissued_at":"2026-07-05T11:38:12.025171Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:12.025171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Watch, Listen, Understand, Mislead: Tri-modal Adversarial Attacks on Short Videos for Content Appropriateness Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mohammed Jawwadul Islam, Montasir Chowdhury, Sadia Ahmmed, Sahid Hossain Mustakim, S M Jishanul Islam, Swakkhar Shatabda, Syed Tasdid Azam Dhrubo, Tashfia Sikder, Ummay Maria Muna","submitted_at":"2025-07-16T07:02:15Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) are increasingly used for content moderation, yet their robustness in short-form video contexts remains underexplored. Current safety evaluations often rely on unimodal attacks, failing to address combined attack vulnerabilities. In this paper, we introduce a comprehensive framework for evaluating the tri-modal safety of MLLMs. First, we present the Short-Video Multimodal Adversarial (SVMA) dataset, comprising diverse short-form videos with human-guided synthetic adversarial attacks. Second, we propose ChimeraBreak, a novel tri-modal attack strategy tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11968","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/2507.11968/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":"2507.11968","created_at":"2026-07-05T11:38:12.025235+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.11968v1","created_at":"2026-07-05T11:38:12.025235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11968","created_at":"2026-07-05T11:38:12.025235+00:00"},{"alias_kind":"pith_short_12","alias_value":"R26JHEFS7GHK","created_at":"2026-07-05T11:38:12.025235+00:00"},{"alias_kind":"pith_short_16","alias_value":"R26JHEFS7GHKYAP4","created_at":"2026-07-05T11:38:12.025235+00:00"},{"alias_kind":"pith_short_8","alias_value":"R26JHEFS","created_at":"2026-07-05T11:38:12.025235+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/R26JHEFS7GHKYAP4BCUFOAGXRF","json":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF.json","graph_json":"https://pith.science/api/pith-number/R26JHEFS7GHKYAP4BCUFOAGXRF/graph.json","events_json":"https://pith.science/api/pith-number/R26JHEFS7GHKYAP4BCUFOAGXRF/events.json","paper":"https://pith.science/paper/R26JHEFS"},"agent_actions":{"view_html":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF","download_json":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF.json","view_paper":"https://pith.science/paper/R26JHEFS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.11968&json=true","fetch_graph":"https://pith.science/api/pith-number/R26JHEFS7GHKYAP4BCUFOAGXRF/graph.json","fetch_events":"https://pith.science/api/pith-number/R26JHEFS7GHKYAP4BCUFOAGXRF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF/action/storage_attestation","attest_author":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF/action/author_attestation","sign_citation":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF/action/citation_signature","submit_replication":"https://pith.science/pith/R26JHEFS7GHKYAP4BCUFOAGXRF/action/replication_record"}},"created_at":"2026-07-05T11:38:12.025235+00:00","updated_at":"2026-07-05T11:38:12.025235+00:00"}