{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DYVW2M3JOQIRXMDXPGOSIW3IJA","short_pith_number":"pith:DYVW2M3J","schema_version":"1.0","canonical_sha256":"1e2b6d336974111bb077799d245b684813bb0775f66af50eadfaa106c0b3c4aa","source":{"kind":"arxiv","id":"2503.04856","version":3},"attestation_state":"computed","paper":{"title":"M2S: Multi-turn to Single-turn jailbreak in Red Teaming for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ashkan Yousefpour, Haon Park, Hyunjun Kim, Junwoo Ha, Sangyoon Yu, Suhyun Kim, Yuna Park","submitted_at":"2025-03-06T07:34:51Z","abstract_excerpt":"We introduce a novel framework for consolidating multi-turn adversarial ``jailbreak'' prompts into single-turn queries, significantly reducing the manual overhead required for adversarial testing of large language models (LLMs). While multi-turn human jailbreaks have been shown to yield high attack success rates, they demand considerable human effort and time. Our multi-turn-to-single-turn (M2S) methods -- Hyphenize, Numberize, and Pythonize -- systematically reformat multi-turn dialogues into structured single-turn prompts. Despite removing iterative back-and-forth interactions, these prompts"},"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":"2503.04856","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T07:34:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0868299efa3df5cbeb1408b5d0f6ea05ddc6e5eaf666d92749106505390c327c","abstract_canon_sha256":"d99199a0a396a8b4b7a0c63811ac3c48fbe596614ab3d18a591b1431fd5a26cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:48:23.400680Z","signature_b64":"mdAjowhtqxnXZKmZDCiWHipfdGlNVrQYEV4BQ+UWaaGbJHfqnjzloD7+N+ZPAEjUngeaVVHXUjApQA6WBfTXDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e2b6d336974111bb077799d245b684813bb0775f66af50eadfaa106c0b3c4aa","last_reissued_at":"2026-07-05T11:48:23.400170Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:48:23.400170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M2S: Multi-turn to Single-turn jailbreak in Red Teaming for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ashkan Yousefpour, Haon Park, Hyunjun Kim, Junwoo Ha, Sangyoon Yu, Suhyun Kim, Yuna Park","submitted_at":"2025-03-06T07:34:51Z","abstract_excerpt":"We introduce a novel framework for consolidating multi-turn adversarial ``jailbreak'' prompts into single-turn queries, significantly reducing the manual overhead required for adversarial testing of large language models (LLMs). While multi-turn human jailbreaks have been shown to yield high attack success rates, they demand considerable human effort and time. Our multi-turn-to-single-turn (M2S) methods -- Hyphenize, Numberize, and Pythonize -- systematically reformat multi-turn dialogues into structured single-turn prompts. Despite removing iterative back-and-forth interactions, these prompts"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.04856","kind":"arxiv","version":3},"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/2503.04856/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":"2503.04856","created_at":"2026-07-05T11:48:23.400240+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.04856v3","created_at":"2026-07-05T11:48:23.400240+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.04856","created_at":"2026-07-05T11:48:23.400240+00:00"},{"alias_kind":"pith_short_12","alias_value":"DYVW2M3JOQIR","created_at":"2026-07-05T11:48:23.400240+00:00"},{"alias_kind":"pith_short_16","alias_value":"DYVW2M3JOQIRXMDX","created_at":"2026-07-05T11:48:23.400240+00:00"},{"alias_kind":"pith_short_8","alias_value":"DYVW2M3J","created_at":"2026-07-05T11:48:23.400240+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/DYVW2M3JOQIRXMDXPGOSIW3IJA","json":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA.json","graph_json":"https://pith.science/api/pith-number/DYVW2M3JOQIRXMDXPGOSIW3IJA/graph.json","events_json":"https://pith.science/api/pith-number/DYVW2M3JOQIRXMDXPGOSIW3IJA/events.json","paper":"https://pith.science/paper/DYVW2M3J"},"agent_actions":{"view_html":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA","download_json":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA.json","view_paper":"https://pith.science/paper/DYVW2M3J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.04856&json=true","fetch_graph":"https://pith.science/api/pith-number/DYVW2M3JOQIRXMDXPGOSIW3IJA/graph.json","fetch_events":"https://pith.science/api/pith-number/DYVW2M3JOQIRXMDXPGOSIW3IJA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA/action/storage_attestation","attest_author":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA/action/author_attestation","sign_citation":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA/action/citation_signature","submit_replication":"https://pith.science/pith/DYVW2M3JOQIRXMDXPGOSIW3IJA/action/replication_record"}},"created_at":"2026-07-05T11:48:23.400240+00:00","updated_at":"2026-07-05T11:48:23.400240+00:00"}