{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SASFCHUZIWW7WG23J5HIRVVYVA","short_pith_number":"pith:SASFCHUZ","schema_version":"1.0","canonical_sha256":"9024511e9945adfb1b5b4f4e88d6b8a83270df6f706e3e997954bca440fe445e","source":{"kind":"arxiv","id":"2502.11054","version":4},"attestation_state":"computed","paper":{"title":"Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Aishan Liu, Dacheng Tao, Deyue Zhang, Quanchen Zou, Siyuan Liang, Xianglong Liu, Xiangzheng Zhang, Yisong Xiao, Zonghao Ying, Zonglei Jing","submitted_at":"2025-02-16T09:27:44Z","abstract_excerpt":"Multi-turn jailbreak attacks simulate real-world human interactions by engaging large language models (LLMs) in iterative dialogues, exposing critical safety vulnerabilities. However, existing methods often struggle to balance semantic coherence with attack effectiveness, resulting in either benign semantic drift or ineffective detection evasion. To address this challenge, we propose Reasoning-Augmented Conversation, a novel multi-turn jailbreak framework that reformulates harmful queries into benign reasoning tasks and leverages LLMs' strong reasoning capabilities to compromise safety alignme"},"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":"2502.11054","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-16T09:27:44Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"248764b954f052be09c96b03f321031f0c28c335ced5155cef476a9a27e50fcf","abstract_canon_sha256":"588da5d21f29b84da7cb8b0d4459629e1495d7a5ef4418e6bd5e575d73e3d785"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:36.072126Z","signature_b64":"8Xt4QyGmlxMw1Wb97P1P5lqEVW8WhdrMZn/kkOUvTQA9xDEWl4u4dtXk6PCkKgYq6BnXTBNHkpxbKpL4PoQVCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9024511e9945adfb1b5b4f4e88d6b8a83270df6f706e3e997954bca440fe445e","last_reissued_at":"2026-07-05T10:28:36.071083Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:36.071083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Aishan Liu, Dacheng Tao, Deyue Zhang, Quanchen Zou, Siyuan Liang, Xianglong Liu, Xiangzheng Zhang, Yisong Xiao, Zonghao Ying, Zonglei Jing","submitted_at":"2025-02-16T09:27:44Z","abstract_excerpt":"Multi-turn jailbreak attacks simulate real-world human interactions by engaging large language models (LLMs) in iterative dialogues, exposing critical safety vulnerabilities. However, existing methods often struggle to balance semantic coherence with attack effectiveness, resulting in either benign semantic drift or ineffective detection evasion. To address this challenge, we propose Reasoning-Augmented Conversation, a novel multi-turn jailbreak framework that reformulates harmful queries into benign reasoning tasks and leverages LLMs' strong reasoning capabilities to compromise safety alignme"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.11054","kind":"arxiv","version":4},"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/2502.11054/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":"2502.11054","created_at":"2026-07-05T10:28:36.071221+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.11054v4","created_at":"2026-07-05T10:28:36.071221+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.11054","created_at":"2026-07-05T10:28:36.071221+00:00"},{"alias_kind":"pith_short_12","alias_value":"SASFCHUZIWW7","created_at":"2026-07-05T10:28:36.071221+00:00"},{"alias_kind":"pith_short_16","alias_value":"SASFCHUZIWW7WG23","created_at":"2026-07-05T10:28:36.071221+00:00"},{"alias_kind":"pith_short_8","alias_value":"SASFCHUZ","created_at":"2026-07-05T10:28:36.071221+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.12710","citing_title":"Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21540","citing_title":"PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2510.10073","citing_title":"SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11002","citing_title":"MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11309","citing_title":"The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04488","citing_title":"A Patch-based Cross-view Regularized Framework for Backdoor Defense in Multimodal Large Language Models","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA","json":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA.json","graph_json":"https://pith.science/api/pith-number/SASFCHUZIWW7WG23J5HIRVVYVA/graph.json","events_json":"https://pith.science/api/pith-number/SASFCHUZIWW7WG23J5HIRVVYVA/events.json","paper":"https://pith.science/paper/SASFCHUZ"},"agent_actions":{"view_html":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA","download_json":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA.json","view_paper":"https://pith.science/paper/SASFCHUZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.11054&json=true","fetch_graph":"https://pith.science/api/pith-number/SASFCHUZIWW7WG23J5HIRVVYVA/graph.json","fetch_events":"https://pith.science/api/pith-number/SASFCHUZIWW7WG23J5HIRVVYVA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA/action/storage_attestation","attest_author":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA/action/author_attestation","sign_citation":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA/action/citation_signature","submit_replication":"https://pith.science/pith/SASFCHUZIWW7WG23J5HIRVVYVA/action/replication_record"}},"created_at":"2026-07-05T10:28:36.071221+00:00","updated_at":"2026-07-05T10:28:36.071221+00:00"}