{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RYA4KMANPWA3U3H4GRS577NZIQ","short_pith_number":"pith:RYA4KMAN","schema_version":"1.0","canonical_sha256":"8e01c5300d7d81ba6cfc3465dffdb94407254a5e9ec0adc3adc44b94f9c12b94","source":{"kind":"arxiv","id":"2505.15406","version":1},"attestation_state":"computed","paper":{"title":"Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chenxi Wang, Guangxian Ouyang, Lang Gao, Mingxuan Cui, Mingzhe Li, Qian Jiang, Xiuying Chen, Yanbo Wang, Zeyu Zhang, Zhenhao Chen, Zirui Song, Zixiang Xu","submitted_at":"2025-05-21T11:47:47Z","abstract_excerpt":"The rise of Large Audio Language Models (LAMs) brings both potential and risks, as their audio outputs may contain harmful or unethical content. However, current research lacks a systematic, quantitative evaluation of LAM safety especially against jailbreak attacks, which are challenging due to the temporal and semantic nature of speech. To bridge this gap, we introduce AJailBench, the first benchmark specifically designed to evaluate jailbreak vulnerabilities in LAMs. We begin by constructing AJailBench-Base, a dataset of 1,495 adversarial audio prompts spanning 10 policy-violating categories"},"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":"2505.15406","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-05-21T11:47:47Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"8d541bebc62956ff420d3e1e4e8b8e938bd971d6228dcfc57983e095a35daa96","abstract_canon_sha256":"b91520660956232fedd748c9f1ebad10e8608bb3684b3119b3cecd238e01e1f5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:45.011269Z","signature_b64":"L4BGgnLfgzorLJ+N2XeeQDOGNt03IV5GZX3+neu74/h6COzQBXwXH04zNTbpsdV18E5CC0GVFDV206cm+GFmAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e01c5300d7d81ba6cfc3465dffdb94407254a5e9ec0adc3adc44b94f9c12b94","last_reissued_at":"2026-07-05T11:06:45.010780Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:45.010780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chenxi Wang, Guangxian Ouyang, Lang Gao, Mingxuan Cui, Mingzhe Li, Qian Jiang, Xiuying Chen, Yanbo Wang, Zeyu Zhang, Zhenhao Chen, Zirui Song, Zixiang Xu","submitted_at":"2025-05-21T11:47:47Z","abstract_excerpt":"The rise of Large Audio Language Models (LAMs) brings both potential and risks, as their audio outputs may contain harmful or unethical content. However, current research lacks a systematic, quantitative evaluation of LAM safety especially against jailbreak attacks, which are challenging due to the temporal and semantic nature of speech. To bridge this gap, we introduce AJailBench, the first benchmark specifically designed to evaluate jailbreak vulnerabilities in LAMs. We begin by constructing AJailBench-Base, a dataset of 1,495 adversarial audio prompts spanning 10 policy-violating categories"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15406","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/2505.15406/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":"2505.15406","created_at":"2026-07-05T11:06:45.010839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15406v1","created_at":"2026-07-05T11:06:45.010839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15406","created_at":"2026-07-05T11:06:45.010839+00:00"},{"alias_kind":"pith_short_12","alias_value":"RYA4KMANPWA3","created_at":"2026-07-05T11:06:45.010839+00:00"},{"alias_kind":"pith_short_16","alias_value":"RYA4KMANPWA3U3H4","created_at":"2026-07-05T11:06:45.010839+00:00"},{"alias_kind":"pith_short_8","alias_value":"RYA4KMAN","created_at":"2026-07-05T11:06:45.010839+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26968","citing_title":"RedVox: Safety and Fairness Gaps in Speech Models Across Languages","ref_index":154,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10581","citing_title":"ParaBridge: Bridging Paralinguistic Perception and Dialogue Behavior in Speech Language Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2505.14226","citing_title":"Phonetic Perturbations Reveal Tokenizer-Rooted Safety Gaps in LLMs","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20266","citing_title":"A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook","ref_index":169,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18168","citing_title":"Acoustic Interference: A New Paradigm Weaponizing Acoustic Latent Semantic for Universal Jailbreak against Large Audio Language Models","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17225","citing_title":"Can Large Audio Language Models Ignore Multilingual Distractors? An Evaluation of Their Selective Auditory Attention Capabilities","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ","json":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ.json","graph_json":"https://pith.science/api/pith-number/RYA4KMANPWA3U3H4GRS577NZIQ/graph.json","events_json":"https://pith.science/api/pith-number/RYA4KMANPWA3U3H4GRS577NZIQ/events.json","paper":"https://pith.science/paper/RYA4KMAN"},"agent_actions":{"view_html":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ","download_json":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ.json","view_paper":"https://pith.science/paper/RYA4KMAN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15406&json=true","fetch_graph":"https://pith.science/api/pith-number/RYA4KMANPWA3U3H4GRS577NZIQ/graph.json","fetch_events":"https://pith.science/api/pith-number/RYA4KMANPWA3U3H4GRS577NZIQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ/action/storage_attestation","attest_author":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ/action/author_attestation","sign_citation":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ/action/citation_signature","submit_replication":"https://pith.science/pith/RYA4KMANPWA3U3H4GRS577NZIQ/action/replication_record"}},"created_at":"2026-07-05T11:06:45.010839+00:00","updated_at":"2026-07-05T11:06:45.010839+00:00"}