{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UJBW5N7IWPZNMRI75HA7QWGICC","short_pith_number":"pith:UJBW5N7I","canonical_record":{"source":{"id":"2608.07813","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T23:23:00Z","cross_cats_sorted":[],"title_canon_sha256":"e33f50912534246ae274eb517a2c5ca5681c8eab8ece0d486fb99235932cca71","abstract_canon_sha256":"9771ef557d83287a57fbb79d922a917030d057dc7ff4afff44c5aa6794d02c1b"},"schema_version":"1.0"},"canonical_sha256":"a2436eb7e8b3f2d6451fe9c1f858c810a0674e87d082fed47cbf5f9569b9f3bb","source":{"kind":"arxiv","id":"2608.07813","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.07813","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"arxiv_version","alias_value":"2608.07813v1","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07813","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_12","alias_value":"UJBW5N7IWPZN","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_16","alias_value":"UJBW5N7IWPZNMRI7","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_8","alias_value":"UJBW5N7I","created_at":"2026-08-11T01:18:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UJBW5N7IWPZNMRI75HA7QWGICC","target":"record","payload":{"canonical_record":{"source":{"id":"2608.07813","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T23:23:00Z","cross_cats_sorted":[],"title_canon_sha256":"e33f50912534246ae274eb517a2c5ca5681c8eab8ece0d486fb99235932cca71","abstract_canon_sha256":"9771ef557d83287a57fbb79d922a917030d057dc7ff4afff44c5aa6794d02c1b"},"schema_version":"1.0"},"canonical_sha256":"a2436eb7e8b3f2d6451fe9c1f858c810a0674e87d082fed47cbf5f9569b9f3bb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T01:18:20.135663Z","signature_b64":"wQysmhQrE4laMAJXDE6lmeWwhBV6DtVFIbN+VbTAvfdRZOITXqcwN+wWGnSHIOH+9Vx+MYWhvRgSAP5zpbBrCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2436eb7e8b3f2d6451fe9c1f858c810a0674e87d082fed47cbf5f9569b9f3bb","last_reissued_at":"2026-08-11T01:18:20.132926Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T01:18:20.132926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.07813","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":"2026-08-11T01:18:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lBxtKL9BvhkdXO67FHkh0tMXf7Iuii3XlootoUiHuo1Zfhx/PgnE4T8zA6ILhHihXa2pOnYhGM2pAmKRDOpWDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:14:49.666600Z"},"content_sha256":"7baceb589e9633c2156629c995d7d75d3ea29b3a5976910a4373a98cc92ab2c8","schema_version":"1.0","event_id":"sha256:7baceb589e9633c2156629c995d7d75d3ea29b3a5976910a4373a98cc92ab2c8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UJBW5N7IWPZNMRI75HA7QWGICC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When the Judge Should Not Decide: Evidence-Locked, Non-Compensatory Selection Bounds LLM-Judge Failure in Reasoning Pipelines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Diksha Goel, Hussain Ahmad, Jun Shen, Shixun Huang, Yiyao Zhang","submitted_at":"2026-08-07T23:23:00Z","abstract_excerpt":"An LLM judge deployed inside a reasoning pipeline does not merely measure quality, it decides which answer ships. We show that the cost of that decision depends less on judge accuracy than on the decision rule the judge is embedded in. On frozen candidate pools from four GRPO policies, an unconstrained scalar DeepSeek-R1-7B judge buys almost nothing over answer-level majority vote (+1.0 pp on 500 GSM8K questions, +0.34 EM on 300 HotpotQA questions), and on a frozen-rule 30-question confirmation split it is 10 points worse than majority, a judge that destroys accuracy while scoring candidates c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07813","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/2608.07813/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":"2026-08-11T01:18:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WWl5arzYtMrC51UXJYs+zhMKB7r1/OQNWZbSs5pJbIXdAOgwCV1yXnoXX40J4APLGyFLShSzREkZQ13Iy49fBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:14:49.667121Z"},"content_sha256":"5c0715fa50c9e86c72adc173db35583c3a36caf34d7cc8da6d25a0c36f07b1a0","schema_version":"1.0","event_id":"sha256:5c0715fa50c9e86c72adc173db35583c3a36caf34d7cc8da6d25a0c36f07b1a0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UJBW5N7IWPZNMRI75HA7QWGICC/bundle.json","state_url":"https://pith.science/pith/UJBW5N7IWPZNMRI75HA7QWGICC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UJBW5N7IWPZNMRI75HA7QWGICC/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":"2026-08-15T21:14:49Z","links":{"resolver":"https://pith.science/pith/UJBW5N7IWPZNMRI75HA7QWGICC","bundle":"https://pith.science/pith/UJBW5N7IWPZNMRI75HA7QWGICC/bundle.json","state":"https://pith.science/pith/UJBW5N7IWPZNMRI75HA7QWGICC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UJBW5N7IWPZNMRI75HA7QWGICC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UJBW5N7IWPZNMRI75HA7QWGICC","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":"9771ef557d83287a57fbb79d922a917030d057dc7ff4afff44c5aa6794d02c1b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T23:23:00Z","title_canon_sha256":"e33f50912534246ae274eb517a2c5ca5681c8eab8ece0d486fb99235932cca71"},"schema_version":"1.0","source":{"id":"2608.07813","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.07813","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"arxiv_version","alias_value":"2608.07813v1","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07813","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_12","alias_value":"UJBW5N7IWPZN","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_16","alias_value":"UJBW5N7IWPZNMRI7","created_at":"2026-08-11T01:18:20Z"},{"alias_kind":"pith_short_8","alias_value":"UJBW5N7I","created_at":"2026-08-11T01:18:20Z"}],"graph_snapshots":[{"event_id":"sha256:5c0715fa50c9e86c72adc173db35583c3a36caf34d7cc8da6d25a0c36f07b1a0","target":"graph","created_at":"2026-08-11T01:18:20Z","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/2608.07813/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"An LLM judge deployed inside a reasoning pipeline does not merely measure quality, it decides which answer ships. We show that the cost of that decision depends less on judge accuracy than on the decision rule the judge is embedded in. On frozen candidate pools from four GRPO policies, an unconstrained scalar DeepSeek-R1-7B judge buys almost nothing over answer-level majority vote (+1.0 pp on 500 GSM8K questions, +0.34 EM on 300 HotpotQA questions), and on a frozen-rule 30-question confirmation split it is 10 points worse than majority, a judge that destroys accuracy while scoring candidates c","authors_text":"Diksha Goel, Hussain Ahmad, Jun Shen, Shixun Huang, Yiyao Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T23:23:00Z","title":"When the Judge Should Not Decide: Evidence-Locked, Non-Compensatory Selection Bounds LLM-Judge Failure in Reasoning Pipelines"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07813","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":"sha256:7baceb589e9633c2156629c995d7d75d3ea29b3a5976910a4373a98cc92ab2c8","target":"record","created_at":"2026-08-11T01:18:20Z","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":"9771ef557d83287a57fbb79d922a917030d057dc7ff4afff44c5aa6794d02c1b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T23:23:00Z","title_canon_sha256":"e33f50912534246ae274eb517a2c5ca5681c8eab8ece0d486fb99235932cca71"},"schema_version":"1.0","source":{"id":"2608.07813","kind":"arxiv","version":1}},"canonical_sha256":"a2436eb7e8b3f2d6451fe9c1f858c810a0674e87d082fed47cbf5f9569b9f3bb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a2436eb7e8b3f2d6451fe9c1f858c810a0674e87d082fed47cbf5f9569b9f3bb","first_computed_at":"2026-08-11T01:18:20.132926Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-11T01:18:20.132926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wQysmhQrE4laMAJXDE6lmeWwhBV6DtVFIbN+VbTAvfdRZOITXqcwN+wWGnSHIOH+9Vx+MYWhvRgSAP5zpbBrCg==","signature_status":"signed_v1","signed_at":"2026-08-11T01:18:20.135663Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.07813","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7baceb589e9633c2156629c995d7d75d3ea29b3a5976910a4373a98cc92ab2c8","sha256:5c0715fa50c9e86c72adc173db35583c3a36caf34d7cc8da6d25a0c36f07b1a0"],"state_sha256":"b4dda4c458af1bf266081a792ffe07e2b2888954f7d4242301a909261cf47cd3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/V3Sv3RbJg+VHsLoa6E0cyGSgUvy0YBUrQ2SxXa+7qEbieYDekqaCu2BDmuYYhgSBYW4kIEHdJ/5dfE67XmMBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T21:14:49.671383Z","bundle_sha256":"d7f15ac312da549a2c530da5c1816beda6a7a426153b132a5504eb81f7ababf9"}}