{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:SZVPL4B26BXSYIVUEJGTRARVK3","short_pith_number":"pith:SZVPL4B2","canonical_record":{"source":{"id":"2306.09983","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-16T17:26:38Z","cross_cats_sorted":["cs.AI","cs.CR","stat.ML"],"title_canon_sha256":"1a8da41745aa07e05f8329a711ba1076bf47ade812898be1c6d731610e2a0a21","abstract_canon_sha256":"b07f8000eb30e426ffcc903ad807b93d55ec67bc463c10874801428e338aabae"},"schema_version":"1.0"},"canonical_sha256":"966af5f03af06f2c22b4224d38823556f3dc67bb117ec24d845cfb8e98047c60","source":{"kind":"arxiv","id":"2306.09983","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.09983","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"arxiv_version","alias_value":"2306.09983v3","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09983","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_12","alias_value":"SZVPL4B26BXS","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_16","alias_value":"SZVPL4B26BXSYIVU","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_8","alias_value":"SZVPL4B2","created_at":"2026-07-05T07:02:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:SZVPL4B26BXSYIVUEJGTRARVK3","target":"record","payload":{"canonical_record":{"source":{"id":"2306.09983","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-16T17:26:38Z","cross_cats_sorted":["cs.AI","cs.CR","stat.ML"],"title_canon_sha256":"1a8da41745aa07e05f8329a711ba1076bf47ade812898be1c6d731610e2a0a21","abstract_canon_sha256":"b07f8000eb30e426ffcc903ad807b93d55ec67bc463c10874801428e338aabae"},"schema_version":"1.0"},"canonical_sha256":"966af5f03af06f2c22b4224d38823556f3dc67bb117ec24d845cfb8e98047c60","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:30.407135Z","signature_b64":"CFP0jL70+ll+lxrSHa0E9wqz0m0KjENUHsdsr/RzjIdjIMHYNxw9IzsxDc+Aoqy5Qeiw/jtKgtguwtqnf4F4DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"966af5f03af06f2c22b4224d38823556f3dc67bb117ec24d845cfb8e98047c60","last_reissued_at":"2026-07-05T07:02:30.406683Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:30.406683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.09983","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:02:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PGGOuZGTXGJwrNo4XRNsVzLBZMLCuE90GeoaXFy0C7qn9GwF02pX3K+5MbP8fwP7IQ/2J5VFyXmfAeUbK0DPDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:37:28.831441Z"},"content_sha256":"52549ae834773575989710af15351df248fb032dfebf5985164101e282f38a5a","schema_version":"1.0","event_id":"sha256:52549ae834773575989710af15351df248fb032dfebf5985164101e282f38a5a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:SZVPL4B26BXSYIVUEJGTRARVK3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Superhuman Models with Consistency Checks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Paleka, Florian Tram\\`er, Lukas Fluri","submitted_at":"2023-06-16T17:26:38Z","abstract_excerpt":"If machine learning models were to achieve superhuman abilities at various reasoning or decision-making tasks, how would we go about evaluating such models, given that humans would necessarily be poor proxies for ground truth? In this paper, we propose a framework for evaluating superhuman models via consistency checks. Our premise is that while the correctness of superhuman decisions may be impossible to evaluate, we can still surface mistakes if the model's decisions fail to satisfy certain logical, human-interpretable rules. We instantiate our framework on three tasks where correctness of d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09983","kind":"arxiv","version":3},"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/2306.09983/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":"2026-07-05T07:02:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CIVRnsCla6gjW8Jfh9jwJalR1AIlVuUzTc8LaiJRXYIXHwidkTDjGJvqBUUxHzsqZlmEQNSwz23bRgecvY+QAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:37:28.831944Z"},"content_sha256":"4cb7e105a556326fb5ca408def3a38e31de3c8a2de334ec51dfdcb923177fe17","schema_version":"1.0","event_id":"sha256:4cb7e105a556326fb5ca408def3a38e31de3c8a2de334ec51dfdcb923177fe17"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SZVPL4B26BXSYIVUEJGTRARVK3/bundle.json","state_url":"https://pith.science/pith/SZVPL4B26BXSYIVUEJGTRARVK3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SZVPL4B26BXSYIVUEJGTRARVK3/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":"2026-08-17T12:37:28Z","links":{"resolver":"https://pith.science/pith/SZVPL4B26BXSYIVUEJGTRARVK3","bundle":"https://pith.science/pith/SZVPL4B26BXSYIVUEJGTRARVK3/bundle.json","state":"https://pith.science/pith/SZVPL4B26BXSYIVUEJGTRARVK3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SZVPL4B26BXSYIVUEJGTRARVK3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SZVPL4B26BXSYIVUEJGTRARVK3","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":"b07f8000eb30e426ffcc903ad807b93d55ec67bc463c10874801428e338aabae","cross_cats_sorted":["cs.AI","cs.CR","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-16T17:26:38Z","title_canon_sha256":"1a8da41745aa07e05f8329a711ba1076bf47ade812898be1c6d731610e2a0a21"},"schema_version":"1.0","source":{"id":"2306.09983","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.09983","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"arxiv_version","alias_value":"2306.09983v3","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09983","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_12","alias_value":"SZVPL4B26BXS","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_16","alias_value":"SZVPL4B26BXSYIVU","created_at":"2026-07-05T07:02:30Z"},{"alias_kind":"pith_short_8","alias_value":"SZVPL4B2","created_at":"2026-07-05T07:02:30Z"}],"graph_snapshots":[{"event_id":"sha256:4cb7e105a556326fb5ca408def3a38e31de3c8a2de334ec51dfdcb923177fe17","target":"graph","created_at":"2026-07-05T07:02:30Z","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/2306.09983/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"If machine learning models were to achieve superhuman abilities at various reasoning or decision-making tasks, how would we go about evaluating such models, given that humans would necessarily be poor proxies for ground truth? In this paper, we propose a framework for evaluating superhuman models via consistency checks. Our premise is that while the correctness of superhuman decisions may be impossible to evaluate, we can still surface mistakes if the model's decisions fail to satisfy certain logical, human-interpretable rules. We instantiate our framework on three tasks where correctness of d","authors_text":"Daniel Paleka, Florian Tram\\`er, Lukas Fluri","cross_cats":["cs.AI","cs.CR","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-16T17:26:38Z","title":"Evaluating Superhuman Models with Consistency Checks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09983","kind":"arxiv","version":3},"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":"sha256:52549ae834773575989710af15351df248fb032dfebf5985164101e282f38a5a","target":"record","created_at":"2026-07-05T07:02:30Z","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":"b07f8000eb30e426ffcc903ad807b93d55ec67bc463c10874801428e338aabae","cross_cats_sorted":["cs.AI","cs.CR","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-16T17:26:38Z","title_canon_sha256":"1a8da41745aa07e05f8329a711ba1076bf47ade812898be1c6d731610e2a0a21"},"schema_version":"1.0","source":{"id":"2306.09983","kind":"arxiv","version":3}},"canonical_sha256":"966af5f03af06f2c22b4224d38823556f3dc67bb117ec24d845cfb8e98047c60","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"966af5f03af06f2c22b4224d38823556f3dc67bb117ec24d845cfb8e98047c60","first_computed_at":"2026-07-05T07:02:30.406683Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:02:30.406683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CFP0jL70+ll+lxrSHa0E9wqz0m0KjENUHsdsr/RzjIdjIMHYNxw9IzsxDc+Aoqy5Qeiw/jtKgtguwtqnf4F4DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:02:30.407135Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.09983","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52549ae834773575989710af15351df248fb032dfebf5985164101e282f38a5a","sha256:4cb7e105a556326fb5ca408def3a38e31de3c8a2de334ec51dfdcb923177fe17"],"state_sha256":"10156afce0b628b3a0de02a0fbc880b5c0826a33cac6cc05c58526fdfd2fb0ff"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KlS/3EO4uHic51c8mz+qZWKFpzvoDGO1/3pKKFGPrUj0aMMOMrDcp2jg2hrpzt5QZdHhnnuQidpqbqggcjbOAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T12:37:28.836802Z","bundle_sha256":"e4a67c8652685fb1b9b89efda6a7e56c5cea9325abc4842d855284950408e3e5"}}