{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:67YERUMPFHQHXP63OGIKMZ35PW","short_pith_number":"pith:67YERUMP","schema_version":"1.0","canonical_sha256":"f7f048d18f29e07bbfdb7190a6677d7db39299e789dcb5d2b2e522cab1558ef4","source":{"kind":"arxiv","id":"2408.11182","version":2},"attestation_state":"computed","paper":{"title":"Hide Your Malicious Goal Into Benign Narratives: Jailbreak Large Language Models through Carrier Articles","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Haizhou Wang, Lan Zhang, Nanqing Luo, Peng Liu, Xiaoyan Sun, Yebo Cao, Zhilong Wang","submitted_at":"2024-08-20T20:35:04Z","abstract_excerpt":"Large Language Model (LLM) jailbreak refers to a type of attack aimed to bypass the safeguard of an LLM to generate contents that are inconsistent with the safe usage guidelines. Based on the insights from the self-attention computation process, this paper proposes a novel blackbox jailbreak approach, which involves crafting the payload prompt by strategically injecting the prohibited query into a carrier article. The carrier article maintains the semantic proximity to the prohibited query, which is automatically produced by combining a hypernymy article and a context, both of which are genera"},"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":"2408.11182","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-08-20T20:35:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1d63f1effa3a0f6015790e599d770cb9a5695543f1bf1a6e779d354c590c8c5d","abstract_canon_sha256":"1bcb9f5ab1bbfeec31fd6eaffb0237c73a8c5703ae099c3d4e7137324b7a6e7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:43.303280Z","signature_b64":"a4OIGGlSIXpGTEMgowfYKdnMqywF+TytE7aqs2IMXMLY3SsmEB4hfs/bROUqON2JPlB3F6jj37jHn248IM9IBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7f048d18f29e07bbfdb7190a6677d7db39299e789dcb5d2b2e522cab1558ef4","last_reissued_at":"2026-07-05T10:10:43.302800Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:43.302800Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hide Your Malicious Goal Into Benign Narratives: Jailbreak Large Language Models through Carrier Articles","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Haizhou Wang, Lan Zhang, Nanqing Luo, Peng Liu, Xiaoyan Sun, Yebo Cao, Zhilong Wang","submitted_at":"2024-08-20T20:35:04Z","abstract_excerpt":"Large Language Model (LLM) jailbreak refers to a type of attack aimed to bypass the safeguard of an LLM to generate contents that are inconsistent with the safe usage guidelines. Based on the insights from the self-attention computation process, this paper proposes a novel blackbox jailbreak approach, which involves crafting the payload prompt by strategically injecting the prohibited query into a carrier article. The carrier article maintains the semantic proximity to the prohibited query, which is automatically produced by combining a hypernymy article and a context, both of which are genera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.11182","kind":"arxiv","version":2},"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/2408.11182/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":"2408.11182","created_at":"2026-07-05T10:10:43.302864+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.11182v2","created_at":"2026-07-05T10:10:43.302864+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.11182","created_at":"2026-07-05T10:10:43.302864+00:00"},{"alias_kind":"pith_short_12","alias_value":"67YERUMPFHQH","created_at":"2026-07-05T10:10:43.302864+00:00"},{"alias_kind":"pith_short_16","alias_value":"67YERUMPFHQHXP63","created_at":"2026-07-05T10:10:43.302864+00:00"},{"alias_kind":"pith_short_8","alias_value":"67YERUMP","created_at":"2026-07-05T10:10:43.302864+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.18487","citing_title":"Adversarial Humanities Benchmark: Results on Stylistic Robustness in Frontier Model Safety","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW","json":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW.json","graph_json":"https://pith.science/api/pith-number/67YERUMPFHQHXP63OGIKMZ35PW/graph.json","events_json":"https://pith.science/api/pith-number/67YERUMPFHQHXP63OGIKMZ35PW/events.json","paper":"https://pith.science/paper/67YERUMP"},"agent_actions":{"view_html":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW","download_json":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW.json","view_paper":"https://pith.science/paper/67YERUMP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.11182&json=true","fetch_graph":"https://pith.science/api/pith-number/67YERUMPFHQHXP63OGIKMZ35PW/graph.json","fetch_events":"https://pith.science/api/pith-number/67YERUMPFHQHXP63OGIKMZ35PW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW/action/storage_attestation","attest_author":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW/action/author_attestation","sign_citation":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW/action/citation_signature","submit_replication":"https://pith.science/pith/67YERUMPFHQHXP63OGIKMZ35PW/action/replication_record"}},"created_at":"2026-07-05T10:10:43.302864+00:00","updated_at":"2026-07-05T10:10:43.302864+00:00"}