{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:JTGXWGS3N42ROGYLFJ6RU4BJEN","short_pith_number":"pith:JTGXWGS3","canonical_record":{"source":{"id":"2607.08065","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-09T02:46:51Z","cross_cats_sorted":[],"title_canon_sha256":"97472542c9c0b281f311a374ecff721d463fa997ff1fd6add09446b224acca11","abstract_canon_sha256":"cbbea24667a8099545dde38f56eb6510e46e89ea58c8f6479f58115b81fd0701"},"schema_version":"1.0"},"canonical_sha256":"4ccd7b1a5b6f35171b0b2a7d1a7029237d12c08e8508eb2979e9bc0d5c3d0960","source":{"kind":"arxiv","id":"2607.08065","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08065","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08065v1","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08065","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_12","alias_value":"JTGXWGS3N42R","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_16","alias_value":"JTGXWGS3N42ROGYL","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_8","alias_value":"JTGXWGS3","created_at":"2026-07-10T00:19:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:JTGXWGS3N42ROGYLFJ6RU4BJEN","target":"record","payload":{"canonical_record":{"source":{"id":"2607.08065","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-09T02:46:51Z","cross_cats_sorted":[],"title_canon_sha256":"97472542c9c0b281f311a374ecff721d463fa997ff1fd6add09446b224acca11","abstract_canon_sha256":"cbbea24667a8099545dde38f56eb6510e46e89ea58c8f6479f58115b81fd0701"},"schema_version":"1.0"},"canonical_sha256":"4ccd7b1a5b6f35171b0b2a7d1a7029237d12c08e8508eb2979e9bc0d5c3d0960","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-10T00:19:09.356571Z","signature_b64":"liLAKAX82VmkYmTUcb8lsOwrSpFzqiuJ0Mzhx67FgX0r1TuAzwwUzV1wa18T2ComcK7/SEhgsKoyLtyEfjy/Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ccd7b1a5b6f35171b0b2a7d1a7029237d12c08e8508eb2979e9bc0d5c3d0960","last_reissued_at":"2026-07-10T00:19:09.356175Z","signature_status":"signed_v1","first_computed_at":"2026-07-10T00:19:09.356175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.08065","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-10T00:19:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IVEjnvd2JlKFGfRYvU++Va6Ug5LdRDnwHm9VI38i+XqVA7nLXQbqxR6Qc0+EGOt9FxKEvNp6ozMpp2jSZkg0Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T08:08:46.437630Z"},"content_sha256":"16558549749ae26a319e8aa7aed655d3ce9072fb333df515fee6607504098f7c","schema_version":"1.0","event_id":"sha256:16558549749ae26a319e8aa7aed655d3ce9072fb333df515fee6607504098f7c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:JTGXWGS3N42ROGYLFJ6RU4BJEN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Kaihua Ding","submitted_at":"2026-07-09T02:46:51Z","abstract_excerpt":"LLM-as-judge (Zheng et al., 2023) is increasingly the default for evaluating AI systems in enterprise pipelines, often scaled to ensembles (Verga et al., 2024) or \"mixture-of-experts\" (Shazeer et al., 2017) panels of judges. These systems share a key assumption: that consistency -- agreement among judges, or among a model's own samples -- indicates correctness. We show this assumption is unreliable. Agreement is not accuracy: a model can agree with itself, and different models can agree with each other, out of shared bias, a memorized heuristic, or an option-position prior rather than truth. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08065","kind":"arxiv","version":1},"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/2607.08065/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":"2026-07-10T00:19:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NC4lQsZeds58vY/YqVP8jvIdUlMI+NZr8CWlQgyhKnPmuULgSUeObwVoLfLcPAreiie7FDcfzmpXBV9n5GU6Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T08:08:46.438048Z"},"content_sha256":"a917a65e9adae7d6a8b6f163ca9f0a764f4661cb12fa0a8455da1eec0ee724b8","schema_version":"1.0","event_id":"sha256:a917a65e9adae7d6a8b6f163ca9f0a764f4661cb12fa0a8455da1eec0ee724b8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/bundle.json","state_url":"https://pith.science/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/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":"2026-07-24T08:08:46Z","links":{"resolver":"https://pith.science/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN","bundle":"https://pith.science/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/bundle.json","state":"https://pith.science/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JTGXWGS3N42ROGYLFJ6RU4BJEN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JTGXWGS3N42ROGYLFJ6RU4BJEN","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":"cbbea24667a8099545dde38f56eb6510e46e89ea58c8f6479f58115b81fd0701","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-09T02:46:51Z","title_canon_sha256":"97472542c9c0b281f311a374ecff721d463fa997ff1fd6add09446b224acca11"},"schema_version":"1.0","source":{"id":"2607.08065","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08065","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08065v1","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08065","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_12","alias_value":"JTGXWGS3N42R","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_16","alias_value":"JTGXWGS3N42ROGYL","created_at":"2026-07-10T00:19:09Z"},{"alias_kind":"pith_short_8","alias_value":"JTGXWGS3","created_at":"2026-07-10T00:19:09Z"}],"graph_snapshots":[{"event_id":"sha256:a917a65e9adae7d6a8b6f163ca9f0a764f4661cb12fa0a8455da1eec0ee724b8","target":"graph","created_at":"2026-07-10T00:19:09Z","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/2607.08065/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-as-judge (Zheng et al., 2023) is increasingly the default for evaluating AI systems in enterprise pipelines, often scaled to ensembles (Verga et al., 2024) or \"mixture-of-experts\" (Shazeer et al., 2017) panels of judges. These systems share a key assumption: that consistency -- agreement among judges, or among a model's own samples -- indicates correctness. We show this assumption is unreliable. Agreement is not accuracy: a model can agree with itself, and different models can agree with each other, out of shared bias, a memorized heuristic, or an option-position prior rather than truth. W","authors_text":"Kaihua Ding","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-09T02:46:51Z","title":"When LLMs Agree, Are They Right? Auditing Self-Consistency and Cross-Model Agreement as Confidence Signals"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08065","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":"sha256:16558549749ae26a319e8aa7aed655d3ce9072fb333df515fee6607504098f7c","target":"record","created_at":"2026-07-10T00:19:09Z","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":"cbbea24667a8099545dde38f56eb6510e46e89ea58c8f6479f58115b81fd0701","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-09T02:46:51Z","title_canon_sha256":"97472542c9c0b281f311a374ecff721d463fa997ff1fd6add09446b224acca11"},"schema_version":"1.0","source":{"id":"2607.08065","kind":"arxiv","version":1}},"canonical_sha256":"4ccd7b1a5b6f35171b0b2a7d1a7029237d12c08e8508eb2979e9bc0d5c3d0960","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4ccd7b1a5b6f35171b0b2a7d1a7029237d12c08e8508eb2979e9bc0d5c3d0960","first_computed_at":"2026-07-10T00:19:09.356175Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T00:19:09.356175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"liLAKAX82VmkYmTUcb8lsOwrSpFzqiuJ0Mzhx67FgX0r1TuAzwwUzV1wa18T2ComcK7/SEhgsKoyLtyEfjy/Bw==","signature_status":"signed_v1","signed_at":"2026-07-10T00:19:09.356571Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.08065","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:16558549749ae26a319e8aa7aed655d3ce9072fb333df515fee6607504098f7c","sha256:a917a65e9adae7d6a8b6f163ca9f0a764f4661cb12fa0a8455da1eec0ee724b8"],"state_sha256":"3ba1edad85804f4d62c5be8e82c45c2b2b223ae2e2ae305a3c6f8635bb11d4e1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TT6I+/mWaKSOhgKZsZ+wcJ0Z6foW7uuqTnxzT7bgT9TH+2UVB254Z2I4kFC5P6ElmTtOF/NRlbl0LnEhCEPJAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-24T08:08:46.440788Z","bundle_sha256":"1983631bf0b386ba1637ed830ed0a62959207ab45f441b3b58bf76ea63312c30"}}