{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SE6CXDTVQW7RCHIYPPJ4X7CHOY","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":"0b5e84caca08889b672c3677244a4751b45513eb3b3a31106a0456f4a4c307bb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-09T14:31:05Z","title_canon_sha256":"125d40a3ea9128e6835674d02451e6e72457ae4ce0e86f59e86a679030ff552d"},"schema_version":"1.0","source":{"id":"2607.08535","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08535","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08535v1","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08535","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_12","alias_value":"SE6CXDTVQW7R","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_16","alias_value":"SE6CXDTVQW7RCHIY","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_8","alias_value":"SE6CXDTV","created_at":"2026-07-10T01:19:53Z"}],"graph_snapshots":[{"event_id":"sha256:40ca32504c5d86cead5543e3d0c8bf37ffc179afb3d33fab78ee4c06588265ae","target":"graph","created_at":"2026-07-10T01:19:53Z","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/2607.08535/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"An LLM-as-judge score can move even when the candidate responses stay fixed, simply because the evaluator has changed. We treat this evaluator-replacement ambiguity as a measurement-validity problem. Across four judgment datasets, we compare two upgrade paths available in practice: scaling Qwen3 dense judges from 1.7B to 32B parameters and moving across MiniMax M2-M2.7 released APIs. The main pattern is that judge upgrades are not interchangeable: only Qwen3 1.7B to 4B gives a robust adjacent gain, while MiniMax adjacent releases do not. Stronger judges reduce but do not remove position and ve","authors_text":"Xiaokun Yang, Yinghan Hou, Zongyou Yang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-09T14:31:05Z","title":"When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08535","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:f2e0e392f878805d67c7f92b36235ad2b25329b161d0a9a9c01d8a385f05f3af","target":"record","created_at":"2026-07-10T01:19:53Z","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":"0b5e84caca08889b672c3677244a4751b45513eb3b3a31106a0456f4a4c307bb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-09T14:31:05Z","title_canon_sha256":"125d40a3ea9128e6835674d02451e6e72457ae4ce0e86f59e86a679030ff552d"},"schema_version":"1.0","source":{"id":"2607.08535","kind":"arxiv","version":1}},"canonical_sha256":"913c2b8e7585bf111d187bd3cbfc47763f8ea2d9853efa34cc9cff866a64d062","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"913c2b8e7585bf111d187bd3cbfc47763f8ea2d9853efa34cc9cff866a64d062","first_computed_at":"2026-07-10T01:19:53.063266Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T01:19:53.063266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OKBSdTB+wHa9DYgNir5afbpKhRYUyyLi0vCLE5tWXv1PeiDeuizOFUo2kGCwtiEeacq6R0qpolAv+1rQmuFRCg==","signature_status":"signed_v1","signed_at":"2026-07-10T01:19:53.063681Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.08535","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2e0e392f878805d67c7f92b36235ad2b25329b161d0a9a9c01d8a385f05f3af","sha256:40ca32504c5d86cead5543e3d0c8bf37ffc179afb3d33fab78ee4c06588265ae"],"state_sha256":"d88157f07e61516f8231a68540b15117488fe8fbfabf346a42b7b1f6a5c3cd41"}