{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:E2S2WS2IWOO3LSCVQV3FIA24S7","short_pith_number":"pith:E2S2WS2I","canonical_record":{"source":{"id":"2505.01420","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T17:57:14Z","cross_cats_sorted":[],"title_canon_sha256":"305f2cdd057ba10b74933f7e8fb4021455d162fc19cd4bb401acdbb11a4e3e62","abstract_canon_sha256":"ec722e5bc4c2a146af77f11c947dda3e2dcaa34f0418b6f1eca346c21499141c"},"schema_version":"1.0"},"canonical_sha256":"26a5ab4b48b39db5c855857654035c97c720a53fddae832eede34bf740c2758b","source":{"kind":"arxiv","id":"2505.01420","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01420","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01420v4","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01420","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_12","alias_value":"E2S2WS2IWOO3","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_16","alias_value":"E2S2WS2IWOO3LSCV","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_8","alias_value":"E2S2WS2I","created_at":"2026-07-05T11:31:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:E2S2WS2IWOO3LSCVQV3FIA24S7","target":"record","payload":{"canonical_record":{"source":{"id":"2505.01420","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T17:57:14Z","cross_cats_sorted":[],"title_canon_sha256":"305f2cdd057ba10b74933f7e8fb4021455d162fc19cd4bb401acdbb11a4e3e62","abstract_canon_sha256":"ec722e5bc4c2a146af77f11c947dda3e2dcaa34f0418b6f1eca346c21499141c"},"schema_version":"1.0"},"canonical_sha256":"26a5ab4b48b39db5c855857654035c97c720a53fddae832eede34bf740c2758b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:27.711359Z","signature_b64":"Xf5g+pA/mKhai7IiJeB9S5KECgjsbZU4Q+3E82E6xVS6ag+ORyeJppKFgi8VsSDW1w0TKlP8sskB45VWf8fABw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26a5ab4b48b39db5c855857654035c97c720a53fddae832eede34bf740c2758b","last_reissued_at":"2026-07-05T11:31:27.710841Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:27.710841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.01420","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:31:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1KxubTQqWfqRBZDjeuNeXRnqhBiEqlJtnavNoRLXfVUsx3ZiQlMihVA+Rk9cEsqyFhn0OFuKTt3B1J4wWdlxBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T14:52:41.693560Z"},"content_sha256":"aabf74b4804d8ff95ea7078aa9a51ac32a1f8fc0f5a85833f20c76fac32f6ac8","schema_version":"1.0","event_id":"sha256:aabf74b4804d8ff95ea7078aa9a51ac32a1f8fc0f5a85833f20c76fac32f6ac8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:E2S2WS2IWOO3LSCVQV3FIA24S7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Frontier Models for Stealth and Situational Awareness","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Allan Dafoe, David Lindner, Lewis Ho, Mary Phuong, Rohin Shah, Roland S. Zimmermann, Sarah Cogan, Victoria Krakovna, Ziyue Wang","submitted_at":"2025-05-02T17:57:14Z","abstract_excerpt":"Recent work has demonstrated the plausibility of frontier AI models scheming -- knowingly and covertly pursuing an objective misaligned with its developer's intentions. Such behavior could be very hard to detect, and if present in future advanced systems, could pose severe loss of control risk. It is therefore important for AI developers to rule out harm from scheming prior to model deployment. In this paper, we present a suite of scheming reasoning evaluations measuring two types of reasoning capabilities that we believe are prerequisites for successful scheming: First, we propose five evalua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01420","kind":"arxiv","version":4},"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/2505.01420/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":"2026-07-05T11:31:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OYtEdya1+QgepcYu2KZ5iF+23jxU7jGXipZizIMQKJZH2uHY2jXmfdjDjKwvF+oGmxY1ZBeOcWSUZrEMJ6Y8Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T14:52:41.693927Z"},"content_sha256":"7bf3148e6a8d5a423d1be390ebbc480768316d09aa481df8da1aab1a25c031f3","schema_version":"1.0","event_id":"sha256:7bf3148e6a8d5a423d1be390ebbc480768316d09aa481df8da1aab1a25c031f3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/bundle.json","state_url":"https://pith.science/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/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":"2026-07-13T14:52:41Z","links":{"resolver":"https://pith.science/pith/E2S2WS2IWOO3LSCVQV3FIA24S7","bundle":"https://pith.science/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/bundle.json","state":"https://pith.science/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/E2S2WS2IWOO3LSCVQV3FIA24S7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:E2S2WS2IWOO3LSCVQV3FIA24S7","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":"ec722e5bc4c2a146af77f11c947dda3e2dcaa34f0418b6f1eca346c21499141c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T17:57:14Z","title_canon_sha256":"305f2cdd057ba10b74933f7e8fb4021455d162fc19cd4bb401acdbb11a4e3e62"},"schema_version":"1.0","source":{"id":"2505.01420","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01420","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01420v4","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01420","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_12","alias_value":"E2S2WS2IWOO3","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_16","alias_value":"E2S2WS2IWOO3LSCV","created_at":"2026-07-05T11:31:27Z"},{"alias_kind":"pith_short_8","alias_value":"E2S2WS2I","created_at":"2026-07-05T11:31:27Z"}],"graph_snapshots":[{"event_id":"sha256:7bf3148e6a8d5a423d1be390ebbc480768316d09aa481df8da1aab1a25c031f3","target":"graph","created_at":"2026-07-05T11:31:27Z","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/2505.01420/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work has demonstrated the plausibility of frontier AI models scheming -- knowingly and covertly pursuing an objective misaligned with its developer's intentions. Such behavior could be very hard to detect, and if present in future advanced systems, could pose severe loss of control risk. It is therefore important for AI developers to rule out harm from scheming prior to model deployment. In this paper, we present a suite of scheming reasoning evaluations measuring two types of reasoning capabilities that we believe are prerequisites for successful scheming: First, we propose five evalua","authors_text":"Allan Dafoe, David Lindner, Lewis Ho, Mary Phuong, Rohin Shah, Roland S. Zimmermann, Sarah Cogan, Victoria Krakovna, Ziyue Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T17:57:14Z","title":"Evaluating Frontier Models for Stealth and Situational Awareness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01420","kind":"arxiv","version":4},"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:aabf74b4804d8ff95ea7078aa9a51ac32a1f8fc0f5a85833f20c76fac32f6ac8","target":"record","created_at":"2026-07-05T11:31:27Z","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":"ec722e5bc4c2a146af77f11c947dda3e2dcaa34f0418b6f1eca346c21499141c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T17:57:14Z","title_canon_sha256":"305f2cdd057ba10b74933f7e8fb4021455d162fc19cd4bb401acdbb11a4e3e62"},"schema_version":"1.0","source":{"id":"2505.01420","kind":"arxiv","version":4}},"canonical_sha256":"26a5ab4b48b39db5c855857654035c97c720a53fddae832eede34bf740c2758b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"26a5ab4b48b39db5c855857654035c97c720a53fddae832eede34bf740c2758b","first_computed_at":"2026-07-05T11:31:27.710841Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:31:27.710841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Xf5g+pA/mKhai7IiJeB9S5KECgjsbZU4Q+3E82E6xVS6ag+ORyeJppKFgi8VsSDW1w0TKlP8sskB45VWf8fABw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:31:27.711359Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.01420","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aabf74b4804d8ff95ea7078aa9a51ac32a1f8fc0f5a85833f20c76fac32f6ac8","sha256:7bf3148e6a8d5a423d1be390ebbc480768316d09aa481df8da1aab1a25c031f3"],"state_sha256":"844ca12963bac89cc7d3a78b0b12bfc351e41ee5384ab8c691b0f06a4cef66c2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rw2duHkAQNnYO4cwbDJ6tiMqH4r1U5gmxTkrTJeox85+/cS8yVwAzAqumyCMFPqzz7qLC1XJM3wjxV7x5C5JBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-13T14:52:41.697331Z","bundle_sha256":"1b5da6637269b48b0c478070e1cf6b1113bc958d7e77799824bdb74bfad363e7"}}