{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YPBJ44ZAB7HKR3STNWBSL7T2VN","short_pith_number":"pith:YPBJ44ZA","schema_version":"1.0","canonical_sha256":"c3c29e73200fcea8ee536d8325fe7aab55ca3f009d56b9085caa25b586045c27","source":{"kind":"arxiv","id":"2505.18864","version":1},"attestation_state":"computed","paper":{"title":"Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Binhao Ma, Hanqing Guo, Rui Duan, Zhengping Jay Luo","submitted_at":"2025-05-24T20:46:36Z","abstract_excerpt":"Recent advances in Multimodal Large Language Models (MLLMs) have significantly enhanced the naturalness and flexibility of human computer interaction by enabling seamless understanding across text, vision, and audio modalities. Among these, voice enabled models such as SpeechGPT have demonstrated considerable improvements in usability, offering expressive, and emotionally responsive interactions that foster deeper connections in real world communication scenarios. However, the use of voice introduces new security risks, as attackers can exploit the unique characteristics of spoken language, su"},"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":"2505.18864","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-24T20:46:36Z","cross_cats_sorted":[],"title_canon_sha256":"e3d33c8ce4a13da152226a7103cad959864c1d5b0918da23ec28ddb8a89f5b6c","abstract_canon_sha256":"ca704c8d02cde7db169c757adfd8ea83f9a37156cc2607ce3bd07fe42cfd6824"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:20.877541Z","signature_b64":"IDQuF1jOYXETv4B0K17i3cHVaGyDJKPKEhKq5xyhiyjCxMh+jtXmtSGVM2XOrjs1M0OZB5sOi0c342ma/wF0Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3c29e73200fcea8ee536d8325fe7aab55ca3f009d56b9085caa25b586045c27","last_reissued_at":"2026-07-05T11:09:20.877026Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:20.877026Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Binhao Ma, Hanqing Guo, Rui Duan, Zhengping Jay Luo","submitted_at":"2025-05-24T20:46:36Z","abstract_excerpt":"Recent advances in Multimodal Large Language Models (MLLMs) have significantly enhanced the naturalness and flexibility of human computer interaction by enabling seamless understanding across text, vision, and audio modalities. Among these, voice enabled models such as SpeechGPT have demonstrated considerable improvements in usability, offering expressive, and emotionally responsive interactions that foster deeper connections in real world communication scenarios. However, the use of voice introduces new security risks, as attackers can exploit the unique characteristics of spoken language, su"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18864","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/2505.18864/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":"2505.18864","created_at":"2026-07-05T11:09:20.877086+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18864v1","created_at":"2026-07-05T11:09:20.877086+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18864","created_at":"2026-07-05T11:09:20.877086+00:00"},{"alias_kind":"pith_short_12","alias_value":"YPBJ44ZAB7HK","created_at":"2026-07-05T11:09:20.877086+00:00"},{"alias_kind":"pith_short_16","alias_value":"YPBJ44ZAB7HKR3ST","created_at":"2026-07-05T11:09:20.877086+00:00"},{"alias_kind":"pith_short_8","alias_value":"YPBJ44ZA","created_at":"2026-07-05T11:09:20.877086+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/YPBJ44ZAB7HKR3STNWBSL7T2VN","json":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN.json","graph_json":"https://pith.science/api/pith-number/YPBJ44ZAB7HKR3STNWBSL7T2VN/graph.json","events_json":"https://pith.science/api/pith-number/YPBJ44ZAB7HKR3STNWBSL7T2VN/events.json","paper":"https://pith.science/paper/YPBJ44ZA"},"agent_actions":{"view_html":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN","download_json":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN.json","view_paper":"https://pith.science/paper/YPBJ44ZA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18864&json=true","fetch_graph":"https://pith.science/api/pith-number/YPBJ44ZAB7HKR3STNWBSL7T2VN/graph.json","fetch_events":"https://pith.science/api/pith-number/YPBJ44ZAB7HKR3STNWBSL7T2VN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN/action/storage_attestation","attest_author":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN/action/author_attestation","sign_citation":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN/action/citation_signature","submit_replication":"https://pith.science/pith/YPBJ44ZAB7HKR3STNWBSL7T2VN/action/replication_record"}},"created_at":"2026-07-05T11:09:20.877086+00:00","updated_at":"2026-07-05T11:09:20.877086+00:00"}