{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:47YF2JLBOUT4X5USJI5JZKQ3NG","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":"bbc6cf5f8c5ce4d36c9bef1e4a8854569b347bb5023a59b35c0b3e93dcbe9be1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-17T11:44:10Z","title_canon_sha256":"d064ab835c52a38c6c42417eedaea7e516ceb9d6f35b57f00461bcdf005411de"},"schema_version":"1.0","source":{"id":"2602.15515","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.15515","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"arxiv_version","alias_value":"2602.15515v2","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.15515","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_12","alias_value":"47YF2JLBOUT4","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_16","alias_value":"47YF2JLBOUT4X5US","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_8","alias_value":"47YF2JLB","created_at":"2026-05-28T01:04:07Z"}],"graph_snapshots":[{"event_id":"sha256:71f8d2c9f0f3c38f6c493ccde6b1fecf427ca6f2006037e5f4b56822ec533a85","target":"graph","created_at":"2026-05-28T01:04:07Z","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/2602.15515/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training against white-box deception detectors has been proposed as a way to make AI systems honest. However, such training risks models learning to obfuscate their deception to evade the detector. Prior work has studied obfuscation only in artificial settings where models were directly rewarded for harmful output. We construct a realistic coding environment where reward hacking via hardcoding test cases naturally occurs, and show that obfuscation emerges in this setting. We introduce a taxonomy of possible outcomes when training against a deception detector. The model either remains honest, o","authors_text":"Adam Gleave, Chris Cundy, Mohammad Taufeeque, Stefan Heimersheim","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-17T11:44:10Z","title":"The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.15515","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:a17fce6bfc2dd8818b98a8222f914713d9412c942c34d37296db09d89683ffbf","target":"record","created_at":"2026-05-28T01:04:07Z","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":"bbc6cf5f8c5ce4d36c9bef1e4a8854569b347bb5023a59b35c0b3e93dcbe9be1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-17T11:44:10Z","title_canon_sha256":"d064ab835c52a38c6c42417eedaea7e516ceb9d6f35b57f00461bcdf005411de"},"schema_version":"1.0","source":{"id":"2602.15515","kind":"arxiv","version":2}},"canonical_sha256":"e7f05d25617527cbf6924a3a9caa1b6985f06d51d2d9c914faba5397cabe008b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e7f05d25617527cbf6924a3a9caa1b6985f06d51d2d9c914faba5397cabe008b","first_computed_at":"2026-05-28T01:04:07.854668Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:07.854668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SElkj7UaGHt7gjYIGbfNs7V0FCzEAonvUCpYb4w/HgfewRqWj3cwV5z8B3uc+Em8V1S3rXqCzShoKrhlhYfJDw==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:07.855243Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.15515","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a17fce6bfc2dd8818b98a8222f914713d9412c942c34d37296db09d89683ffbf","sha256:71f8d2c9f0f3c38f6c493ccde6b1fecf427ca6f2006037e5f4b56822ec533a85"],"state_sha256":"77b42a5b0e59d14dd905f1c36524ebb570afbe700722e7d6633320ad3cab0087"}