{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BEGOZ5TPMQBUQOMR6WSU4OLVOO","short_pith_number":"pith:BEGOZ5TP","canonical_record":{"source":{"id":"2506.13206","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T08:10:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"93b1cbd8aefa7412c1ad18849fadc2a0ffd49511213df2c486dc47048ad97acc","abstract_canon_sha256":"aba7990f27950533650893ff4b31b131870ec442799bcdbc08cddec6b59dfb14"},"schema_version":"1.0"},"canonical_sha256":"090cecf66f6403483991f5a54e3975738bee159e2c001b7d9f1367c0c95af098","source":{"kind":"arxiv","id":"2506.13206","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.13206","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.13206v2","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13206","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_12","alias_value":"BEGOZ5TPMQBU","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_16","alias_value":"BEGOZ5TPMQBUQOMR","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_8","alias_value":"BEGOZ5TP","created_at":"2026-07-05T11:34:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BEGOZ5TPMQBUQOMR6WSU4OLVOO","target":"record","payload":{"canonical_record":{"source":{"id":"2506.13206","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T08:10:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"93b1cbd8aefa7412c1ad18849fadc2a0ffd49511213df2c486dc47048ad97acc","abstract_canon_sha256":"aba7990f27950533650893ff4b31b131870ec442799bcdbc08cddec6b59dfb14"},"schema_version":"1.0"},"canonical_sha256":"090cecf66f6403483991f5a54e3975738bee159e2c001b7d9f1367c0c95af098","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:34:52.137267Z","signature_b64":"PQNQeGsAcVdhZSgtBTquSLOFQg9C2L0xt7Aw1q3g6qZJE6BVc0YPKTeFjJj2Fd++H8QHJlbZvZcHNkt/Eu+XBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"090cecf66f6403483991f5a54e3975738bee159e2c001b7d9f1367c0c95af098","last_reissued_at":"2026-07-05T11:34:52.136735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:34:52.136735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.13206","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:34:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YGf0/hnnUiT5jNmgzgV+VnD1W2APmTaR7+d6CvqpWDql94lumr6hC4HXf2njzYLwI8oLaGuoaO7OnAPHIG+kAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:17:05.718711Z"},"content_sha256":"bfbea818033fd3b00925f727c9d62d1dccaea86f86ce5d66c464ed0a40f1fbb9","schema_version":"1.0","event_id":"sha256:bfbea818033fd3b00925f727c9d62d1dccaea86f86ce5d66c464ed0a40f1fbb9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BEGOZ5TPMQBUQOMR6WSU4OLVOO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"James Chua, Jan Betley, Mia Taylor, Owain Evans","submitted_at":"2025-06-16T08:10:04Z","abstract_excerpt":"Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain (e.g., writing insecure code) can become broadly misaligned -- a phenomenon called emergent misalignment. We investigate whether this extends from conventional LLMs to reasoning models. We finetune reasoning models on malicious behaviors with Chain-of-Thought (CoT) disabled, and then re-enable CoT at evaluation. Like conventional LLMs, reasoning models become broadly misaligned. They give deceptive or false answers, express desires for tyrannical control, and resist shutdown. Inspecting the CoT preceding these misal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13206","kind":"arxiv","version":2},"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/2506.13206/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:34:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kOolkwX/rQO1w3oGm565Zhvxa2JpXibFIqCx9Bv+P3oqSFupOvYd4/TIjpz4HYHdpv25GuaRYxv06HM9jjcdCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:17:05.719245Z"},"content_sha256":"f1a73ae7b8c9d9a47b3e673324bcc95a58141ec6e8907fd8e00b092de1b5699b","schema_version":"1.0","event_id":"sha256:f1a73ae7b8c9d9a47b3e673324bcc95a58141ec6e8907fd8e00b092de1b5699b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/bundle.json","state_url":"https://pith.science/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T18:17:05Z","links":{"resolver":"https://pith.science/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO","bundle":"https://pith.science/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/bundle.json","state":"https://pith.science/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BEGOZ5TPMQBUQOMR6WSU4OLVOO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BEGOZ5TPMQBUQOMR6WSU4OLVOO","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":"aba7990f27950533650893ff4b31b131870ec442799bcdbc08cddec6b59dfb14","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T08:10:04Z","title_canon_sha256":"93b1cbd8aefa7412c1ad18849fadc2a0ffd49511213df2c486dc47048ad97acc"},"schema_version":"1.0","source":{"id":"2506.13206","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.13206","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.13206v2","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13206","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_12","alias_value":"BEGOZ5TPMQBU","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_16","alias_value":"BEGOZ5TPMQBUQOMR","created_at":"2026-07-05T11:34:52Z"},{"alias_kind":"pith_short_8","alias_value":"BEGOZ5TP","created_at":"2026-07-05T11:34:52Z"}],"graph_snapshots":[{"event_id":"sha256:f1a73ae7b8c9d9a47b3e673324bcc95a58141ec6e8907fd8e00b092de1b5699b","target":"graph","created_at":"2026-07-05T11:34:52Z","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/2506.13206/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain (e.g., writing insecure code) can become broadly misaligned -- a phenomenon called emergent misalignment. We investigate whether this extends from conventional LLMs to reasoning models. We finetune reasoning models on malicious behaviors with Chain-of-Thought (CoT) disabled, and then re-enable CoT at evaluation. Like conventional LLMs, reasoning models become broadly misaligned. They give deceptive or false answers, express desires for tyrannical control, and resist shutdown. Inspecting the CoT preceding these misal","authors_text":"James Chua, Jan Betley, Mia Taylor, Owain Evans","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T08:10:04Z","title":"Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13206","kind":"arxiv","version":2},"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:bfbea818033fd3b00925f727c9d62d1dccaea86f86ce5d66c464ed0a40f1fbb9","target":"record","created_at":"2026-07-05T11:34:52Z","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":"aba7990f27950533650893ff4b31b131870ec442799bcdbc08cddec6b59dfb14","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-16T08:10:04Z","title_canon_sha256":"93b1cbd8aefa7412c1ad18849fadc2a0ffd49511213df2c486dc47048ad97acc"},"schema_version":"1.0","source":{"id":"2506.13206","kind":"arxiv","version":2}},"canonical_sha256":"090cecf66f6403483991f5a54e3975738bee159e2c001b7d9f1367c0c95af098","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"090cecf66f6403483991f5a54e3975738bee159e2c001b7d9f1367c0c95af098","first_computed_at":"2026-07-05T11:34:52.136735Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:34:52.136735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PQNQeGsAcVdhZSgtBTquSLOFQg9C2L0xt7Aw1q3g6qZJE6BVc0YPKTeFjJj2Fd++H8QHJlbZvZcHNkt/Eu+XBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:34:52.137267Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.13206","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bfbea818033fd3b00925f727c9d62d1dccaea86f86ce5d66c464ed0a40f1fbb9","sha256:f1a73ae7b8c9d9a47b3e673324bcc95a58141ec6e8907fd8e00b092de1b5699b"],"state_sha256":"86a34687ede03937339adf6102dd44eef4f00a79d398804a02c4bd4637cd614f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hpED+ma9fHs+QZDbUcxHBa4EvGFGxhmCKW5gFlIDXFikugPwyOo6MrhHtMZnLuoYCN+bDGVKGHLR8V5pGrEaAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T18:17:05.724702Z","bundle_sha256":"d190b9b6f3ba49be29240b50fef163045041b2f9365d2ab5ee9bddcee3f67a7b"}}