{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:P7ZXU3BGC2QIWEVTIG7TH5H3M4","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":"2de3a90ce968483a77569ff123d4f2f75304dc9847845ec144eab57295494623","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T09:41:15Z","title_canon_sha256":"0f113d3f5890f2fdd9e42a583de69dca32d260fcdfaf726a402dd5ee27d2f005"},"schema_version":"1.0","source":{"id":"2605.26772","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.26772","created_at":"2026-05-27T01:06:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.26772v1","created_at":"2026-05-27T01:06:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26772","created_at":"2026-05-27T01:06:11Z"},{"alias_kind":"pith_short_12","alias_value":"P7ZXU3BGC2QI","created_at":"2026-05-27T01:06:11Z"},{"alias_kind":"pith_short_16","alias_value":"P7ZXU3BGC2QIWEVT","created_at":"2026-05-27T01:06:11Z"},{"alias_kind":"pith_short_8","alias_value":"P7ZXU3BG","created_at":"2026-05-27T01:06:11Z"}],"graph_snapshots":[{"event_id":"sha256:596e5abcb748baeb45ed9538650830aac8831c0eabada4c9e1b58c25aa345c03","target":"graph","created_at":"2026-05-27T01:06:11Z","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/2605.26772/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large reasoning models (LRMs) generate chain-of-thought (CoT) traces before producing final outputs, introducing a dynamic internal state that may complicate control mechanisms such as refusal. Unlike instruction-tuned LLMs, where refusal is mediated by a single directional subspace, refusal in large reasoning models (LRMs) additionally depends on the CoT. In DeepSeek-R1-Distill-LLaMA-8B, activation steering reverses refusal in only 39% of cases when the CoT is kept fixed, but removing the CoT entirely increases this to 70%, indicating that the CoT actively reinforces refusal. In a two-stage i","authors_text":"Bela Gipp, Dominik Meier, Jiachen Zhao, Kia-J\\\"ung Yang, Terry Ruas","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T09:41:15Z","title":"Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26772","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:e75c9f67ea5986f1097e93ea4b2af264f808d22d335048778ff952e779516938","target":"record","created_at":"2026-05-27T01:06:11Z","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":"2de3a90ce968483a77569ff123d4f2f75304dc9847845ec144eab57295494623","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T09:41:15Z","title_canon_sha256":"0f113d3f5890f2fdd9e42a583de69dca32d260fcdfaf726a402dd5ee27d2f005"},"schema_version":"1.0","source":{"id":"2605.26772","kind":"arxiv","version":1}},"canonical_sha256":"7ff37a6c2616a08b12b341bf33f4fb67205df8c26282b5d941eb47175a06f53a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7ff37a6c2616a08b12b341bf33f4fb67205df8c26282b5d941eb47175a06f53a","first_computed_at":"2026-05-27T01:06:11.870565Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:06:11.870565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JqY3xbibl6QBnbL9/zjYaczeyQnVI5KN8Xjh9jJby4zn1iDFht7t7wLW2t5W2NytFbtgmF/mKxxM6c4lat26Bg==","signature_status":"signed_v1","signed_at":"2026-05-27T01:06:11.875328Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.26772","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e75c9f67ea5986f1097e93ea4b2af264f808d22d335048778ff952e779516938","sha256:596e5abcb748baeb45ed9538650830aac8831c0eabada4c9e1b58c25aa345c03"],"state_sha256":"e28b4ea8cb1f0138e8ad5d7806da05bbdab7c6a3ffd11090a8a8971446828bda"}