{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:H7OYBQRSRJGM5QX4YRAAGCBROZ","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":"deaaad2e03cc8336c48bb411cfa2bc5a41465094a35a3b4525454ab2d2c4de0e","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T11:50:28Z","title_canon_sha256":"53483d04e46e896127cc2cd81681cd526981d530f3934423129ac68cbe2f46b1"},"schema_version":"1.0","source":{"id":"2606.10740","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.10740","created_at":"2026-06-10T01:10:37Z"},{"alias_kind":"arxiv_version","alias_value":"2606.10740v1","created_at":"2026-06-10T01:10:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10740","created_at":"2026-06-10T01:10:37Z"},{"alias_kind":"pith_short_12","alias_value":"H7OYBQRSRJGM","created_at":"2026-06-10T01:10:37Z"},{"alias_kind":"pith_short_16","alias_value":"H7OYBQRSRJGM5QX4","created_at":"2026-06-10T01:10:37Z"},{"alias_kind":"pith_short_8","alias_value":"H7OYBQRS","created_at":"2026-06-10T01:10:37Z"}],"graph_snapshots":[{"event_id":"sha256:18b00ea6fe6e5f4aa368e75feab6315b1bf2d00a60fbf9010c1944edc49eab13","target":"graph","created_at":"2026-06-10T01:10:37Z","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/2606.10740/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Failures in multi-turn reasoning models are largely invisible to terminal-score evaluation. A model can lock onto an unsafe stance early in a long dialogue, yet its final-turn refusal rate may appear indistinguishable from a robustly aligned baseline. To expose these hidden temporal dynamics, we propose a trace-level diagnostic - the CoT-Output 2x2 safety matrix. This framework labels every turn along two independent axes (internal reasoning and visible output), yielding four operationally defined failure cells: robust alignment, alignment faking, overt jailbreak, and a distinct failure mode w","authors_text":"Nils Lukas, Sai Kartheek Reddy Kasu, Samuele Poppi","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T11:50:28Z","title":"When the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10740","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":"sha256:6a87c5d5698694138c430382c2e2adde1ab0632faf39d60550d92775617eb016","target":"record","created_at":"2026-06-10T01:10:37Z","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":"deaaad2e03cc8336c48bb411cfa2bc5a41465094a35a3b4525454ab2d2c4de0e","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T11:50:28Z","title_canon_sha256":"53483d04e46e896127cc2cd81681cd526981d530f3934423129ac68cbe2f46b1"},"schema_version":"1.0","source":{"id":"2606.10740","kind":"arxiv","version":1}},"canonical_sha256":"3fdd80c2328a4ccec2fcc4400308317651c81cfd38116815cc23b621cd9038e4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3fdd80c2328a4ccec2fcc4400308317651c81cfd38116815cc23b621cd9038e4","first_computed_at":"2026-06-10T01:10:37.691811Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T01:10:37.691811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"f4LH0hQ4bmyzt/QBvsdd9DqfXl26IfOH/koM6xV/jN/y61lH5z67roPUvZA04vsNSn2/TyB2tGY8k3LLeNaeCg==","signature_status":"signed_v1","signed_at":"2026-06-10T01:10:37.692718Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.10740","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6a87c5d5698694138c430382c2e2adde1ab0632faf39d60550d92775617eb016","sha256:18b00ea6fe6e5f4aa368e75feab6315b1bf2d00a60fbf9010c1944edc49eab13"],"state_sha256":"e54536eab10660cc7d9546eb69a1e99bb25c2f622a41c0f32d9f55f59d2c64f0"}