{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LYJH7TDII7HHDXDSMDO4AVAYDZ","short_pith_number":"pith:LYJH7TDI","schema_version":"1.0","canonical_sha256":"5e127fcc6847ce71dc7260ddc054181e7a0838b20c6e0c7320d522f595b34080","source":{"kind":"arxiv","id":"2411.11407","version":1},"attestation_state":"computed","paper":{"title":"The Dark Side of Trust: Authority Citation-Driven Jailbreak Attacks on Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jizhong Han, Songlin Hu, Xikang Yang, Xuehai Tang","submitted_at":"2024-11-18T09:28:58Z","abstract_excerpt":"The widespread deployment of large language models (LLMs) across various domains has showcased their immense potential while exposing significant safety vulnerabilities. A major concern is ensuring that LLM-generated content aligns with human values. Existing jailbreak techniques reveal how this alignment can be compromised through specific prompts or adversarial suffixes. In this study, we introduce a new threat: LLMs' bias toward authority. While this inherent bias can improve the quality of outputs generated by LLMs, it also introduces a potential vulnerability, increasing the risk of produ"},"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":"2411.11407","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T09:28:58Z","cross_cats_sorted":[],"title_canon_sha256":"3a959cf138fe1c20c3c9a5f85df73441983f07ceac8fc8f76af031b24e480c93","abstract_canon_sha256":"8b61efadb74177eaddc56150b6f24e03c330df9e67e6ed78541eab2c495341cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:36:55.903624Z","signature_b64":"aUbA61QTiPzkOjGjGCXO20Z0/4eZs4SlGHorg92z2XYAtczGtUGenHnxrxaG3AL8AO4xqixxJ7qcTke3QbP3BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e127fcc6847ce71dc7260ddc054181e7a0838b20c6e0c7320d522f595b34080","last_reissued_at":"2026-07-05T09:36:55.903168Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:36:55.903168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Dark Side of Trust: Authority Citation-Driven Jailbreak Attacks on Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jizhong Han, Songlin Hu, Xikang Yang, Xuehai Tang","submitted_at":"2024-11-18T09:28:58Z","abstract_excerpt":"The widespread deployment of large language models (LLMs) across various domains has showcased their immense potential while exposing significant safety vulnerabilities. A major concern is ensuring that LLM-generated content aligns with human values. Existing jailbreak techniques reveal how this alignment can be compromised through specific prompts or adversarial suffixes. In this study, we introduce a new threat: LLMs' bias toward authority. While this inherent bias can improve the quality of outputs generated by LLMs, it also introduces a potential vulnerability, increasing the risk of produ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.11407","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/2411.11407/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":"2411.11407","created_at":"2026-07-05T09:36:55.903220+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.11407v1","created_at":"2026-07-05T09:36:55.903220+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.11407","created_at":"2026-07-05T09:36:55.903220+00:00"},{"alias_kind":"pith_short_12","alias_value":"LYJH7TDII7HH","created_at":"2026-07-05T09:36:55.903220+00:00"},{"alias_kind":"pith_short_16","alias_value":"LYJH7TDII7HHDXDS","created_at":"2026-07-05T09:36:55.903220+00:00"},{"alias_kind":"pith_short_8","alias_value":"LYJH7TDI","created_at":"2026-07-05T09:36:55.903220+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07461","citing_title":"Mitigating Taint-Style Vulnerabilities in MCP Servers via Security-Aware Tool Descriptions","ref_index":61,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07709","citing_title":"IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ","json":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ.json","graph_json":"https://pith.science/api/pith-number/LYJH7TDII7HHDXDSMDO4AVAYDZ/graph.json","events_json":"https://pith.science/api/pith-number/LYJH7TDII7HHDXDSMDO4AVAYDZ/events.json","paper":"https://pith.science/paper/LYJH7TDI"},"agent_actions":{"view_html":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ","download_json":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ.json","view_paper":"https://pith.science/paper/LYJH7TDI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.11407&json=true","fetch_graph":"https://pith.science/api/pith-number/LYJH7TDII7HHDXDSMDO4AVAYDZ/graph.json","fetch_events":"https://pith.science/api/pith-number/LYJH7TDII7HHDXDSMDO4AVAYDZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ/action/storage_attestation","attest_author":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ/action/author_attestation","sign_citation":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ/action/citation_signature","submit_replication":"https://pith.science/pith/LYJH7TDII7HHDXDSMDO4AVAYDZ/action/replication_record"}},"created_at":"2026-07-05T09:36:55.903220+00:00","updated_at":"2026-07-05T09:36:55.903220+00:00"}