{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3CDPA7ZZGQEM3AYDLEO6ZBHI6J","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"360dc3bdea8c84f22f9ba805967ad469ef649d1e407c8723455632849c1172f7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-31T18:38:23Z","title_canon_sha256":"037cbf5dc339102b4004d283326986d0c5b023743894c0de95b147789ed1f02d"},"schema_version":"1.0","source":{"id":"2506.00668","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00668","created_at":"2026-07-05T11:13:38Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00668v1","created_at":"2026-07-05T11:13:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00668","created_at":"2026-07-05T11:13:38Z"},{"alias_kind":"pith_short_12","alias_value":"3CDPA7ZZGQEM","created_at":"2026-07-05T11:13:38Z"},{"alias_kind":"pith_short_16","alias_value":"3CDPA7ZZGQEM3AYD","created_at":"2026-07-05T11:13:38Z"},{"alias_kind":"pith_short_8","alias_value":"3CDPA7ZZ","created_at":"2026-07-05T11:13:38Z"}],"graph_snapshots":[{"event_id":"sha256:64e39ebadb68a5e65d01c5b2b1c06524e4dca0fb276161bb03cc36d9cf72e6de","target":"graph","created_at":"2026-07-05T11:13:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.00668/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Malicious attackers can exploit large language models (LLMs) by engaging them in multi-turn dialogues to achieve harmful objectives, posing significant safety risks to society. To address this challenge, we propose a novel defense mechanism: SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues (STREAM). STREAM defends LLMs against multi-turn attacks while preserving their functional capabilities. Our approach involves constructing a human-annotated dataset, the Safety Reasoning Multi-turn Dialogues dataset, which is used to fine-tune a plug-and-play safety reasoning moderator. This ","authors_text":"Amin Hass, Aolin Ding, Benjamin F Morris, Bhavna Gopal, Changwei Liu, Hai Li, Isaac Jacobson, James Kiessling, Jianyi Zhang, Louis DiValentin, Martin Kuo, Maziyar Baran Pouyan, Nicolas Ramos, Randolph Linderman, Yiran Chen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-31T18:38:23Z","title":"SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00668","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ee8383242f9a29a4f20f3fe2166823fd17ea94fa9a49bb3a35664fdcc80cb245","target":"record","created_at":"2026-07-05T11:13:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"360dc3bdea8c84f22f9ba805967ad469ef649d1e407c8723455632849c1172f7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-31T18:38:23Z","title_canon_sha256":"037cbf5dc339102b4004d283326986d0c5b023743894c0de95b147789ed1f02d"},"schema_version":"1.0","source":{"id":"2506.00668","kind":"arxiv","version":1}},"canonical_sha256":"d886f07f393408cd8303591dec84e8f266956bf7f1406d4f75bf3cb861f6aa22","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d886f07f393408cd8303591dec84e8f266956bf7f1406d4f75bf3cb861f6aa22","first_computed_at":"2026-07-05T11:13:38.357390Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:38.357390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"71uSyalmZQDsCIAVUscXhsvMRKlmnkiXXol2p8QHXAP0yRdgVmnzcmW0OozLiYHpJbl+uJggi8IKzPqph8KBDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:38.357941Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.00668","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ee8383242f9a29a4f20f3fe2166823fd17ea94fa9a49bb3a35664fdcc80cb245","sha256:64e39ebadb68a5e65d01c5b2b1c06524e4dca0fb276161bb03cc36d9cf72e6de"],"state_sha256":"f49b016b3a80ce5a80e380cf06f6d57835c11c4562b1abe7d41296e1d97ba178"}