{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VZUH6S2KPRUXTSSYC2AXHIXC3L","short_pith_number":"pith:VZUH6S2K","schema_version":"1.0","canonical_sha256":"ae687f4b4a7c6979ca58168173a2e2daf9f076fdbab5086b0439e352a13c151b","source":{"kind":"arxiv","id":"2501.09039","version":1},"attestation_state":"computed","paper":{"title":"Playing Devil's Advocate: Unmasking Toxicity and Vulnerabilities in Large Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CR","authors_text":"Abdulkadir Erol, Agnik Saha, Mehmet Emin Aktas, Trilok Padhi, Ugur Kursuncu","submitted_at":"2025-01-14T21:27:40Z","abstract_excerpt":"The rapid advancement of Large Vision-Language Models (LVLMs) has enhanced capabilities offering potential applications from content creation to productivity enhancement. Despite their innovative potential, LVLMs exhibit vulnerabilities, especially in generating potentially toxic or unsafe responses. Malicious actors can exploit these vulnerabilities to propagate toxic content in an automated (or semi-) manner, leveraging the susceptibility of LVLMs to deception via strategically crafted prompts without fine-tuning or compute-intensive procedures. Despite the red-teaming efforts and inherent p"},"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":"2501.09039","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-14T21:27:40Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"4fbb772a8f938db60cf2e96a82ba9d40e1168e2ecda898028dd8fe74d0397cd9","abstract_canon_sha256":"9ac1a02acf81ac80c501c29244c7b3a16c28f54f7aef4a0e7ebc293c8e24e45c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:01:40.760762Z","signature_b64":"8EsQ/RMMpuXpBFKwoqcbnVuGK8QeT3GGboDfAv/da+SKOikFl2OXbUf3KwD29G/JOdAKOgnZQGFjP0ML09FBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae687f4b4a7c6979ca58168173a2e2daf9f076fdbab5086b0439e352a13c151b","last_reissued_at":"2026-07-05T10:01:40.760241Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:01:40.760241Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Playing Devil's Advocate: Unmasking Toxicity and Vulnerabilities in Large Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CR","authors_text":"Abdulkadir Erol, Agnik Saha, Mehmet Emin Aktas, Trilok Padhi, Ugur Kursuncu","submitted_at":"2025-01-14T21:27:40Z","abstract_excerpt":"The rapid advancement of Large Vision-Language Models (LVLMs) has enhanced capabilities offering potential applications from content creation to productivity enhancement. Despite their innovative potential, LVLMs exhibit vulnerabilities, especially in generating potentially toxic or unsafe responses. Malicious actors can exploit these vulnerabilities to propagate toxic content in an automated (or semi-) manner, leveraging the susceptibility of LVLMs to deception via strategically crafted prompts without fine-tuning or compute-intensive procedures. Despite the red-teaming efforts and inherent p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09039","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/2501.09039/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":"2501.09039","created_at":"2026-07-05T10:01:40.760308+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.09039v1","created_at":"2026-07-05T10:01:40.760308+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09039","created_at":"2026-07-05T10:01:40.760308+00:00"},{"alias_kind":"pith_short_12","alias_value":"VZUH6S2KPRUX","created_at":"2026-07-05T10:01:40.760308+00:00"},{"alias_kind":"pith_short_16","alias_value":"VZUH6S2KPRUXTSSY","created_at":"2026-07-05T10:01:40.760308+00:00"},{"alias_kind":"pith_short_8","alias_value":"VZUH6S2K","created_at":"2026-07-05T10:01:40.760308+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2512.05929","citing_title":"LLM Harms: A Taxonomy and Discussion","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L","json":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L.json","graph_json":"https://pith.science/api/pith-number/VZUH6S2KPRUXTSSYC2AXHIXC3L/graph.json","events_json":"https://pith.science/api/pith-number/VZUH6S2KPRUXTSSYC2AXHIXC3L/events.json","paper":"https://pith.science/paper/VZUH6S2K"},"agent_actions":{"view_html":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L","download_json":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L.json","view_paper":"https://pith.science/paper/VZUH6S2K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.09039&json=true","fetch_graph":"https://pith.science/api/pith-number/VZUH6S2KPRUXTSSYC2AXHIXC3L/graph.json","fetch_events":"https://pith.science/api/pith-number/VZUH6S2KPRUXTSSYC2AXHIXC3L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L/action/storage_attestation","attest_author":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L/action/author_attestation","sign_citation":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L/action/citation_signature","submit_replication":"https://pith.science/pith/VZUH6S2KPRUXTSSYC2AXHIXC3L/action/replication_record"}},"created_at":"2026-07-05T10:01:40.760308+00:00","updated_at":"2026-07-05T10:01:40.760308+00:00"}