{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NFBBO4TL7SYOS7KMX4PRZ2RYU7","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":"706fa7fd9d3a60be08ece1f9d00b9c0e2fa2350e433e1c69eeb9e7895fb3b74a","cross_cats_sorted":["cs.HC","physics.data-an"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T07:31:44Z","title_canon_sha256":"78e1cae135e4185ba867359a42d4e2e863d2184fd6bfe9e18c49eebba86d9ea1"},"schema_version":"1.0","source":{"id":"2606.00093","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00093","created_at":"2026-06-02T00:03:14Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00093v1","created_at":"2026-06-02T00:03:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00093","created_at":"2026-06-02T00:03:14Z"},{"alias_kind":"pith_short_12","alias_value":"NFBBO4TL7SYO","created_at":"2026-06-02T00:03:14Z"},{"alias_kind":"pith_short_16","alias_value":"NFBBO4TL7SYOS7KM","created_at":"2026-06-02T00:03:14Z"},{"alias_kind":"pith_short_8","alias_value":"NFBBO4TL","created_at":"2026-06-02T00:03:14Z"}],"graph_snapshots":[{"event_id":"sha256:f9766f5c422b1b3aca5b4a503ea75279b7065818c2fe04d474975004f9f2005a","target":"graph","created_at":"2026-06-02T00:03:14Z","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/2606.00093/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Validating an LLM judge against human annotations usually means reporting several agreement statistics: accuracy, precision, recall, $F_1$, Cohen's $\\kappa$, and one or more rank correlations. A survey of 24 recent LLM-as-judge papers finds metric choice entangled with the judgment scale, tie handling, invalid outputs, and abstention handling, and those choices rarely stated. For binary criteria -- the common case in rubric-based evaluation, where each criterion is graded MET or UNMET -- most of the reported numbers are redundant: Pearson's $r$, Spearman's $\\rho$, Kendall's $\\tau_b$, the phi c","authors_text":"Chris Callison-Burch, Delip Rao","cross_cats":["cs.HC","physics.data-an"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T07:31:44Z","title":"Agreement Metrics for LLM-as-Judge Evaluation: What to Report and Why"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00093","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:f5f3dc2109ac2f714068fbd6c6c543da28a85227df4ead4249e52917872c46e0","target":"record","created_at":"2026-06-02T00:03:14Z","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":"706fa7fd9d3a60be08ece1f9d00b9c0e2fa2350e433e1c69eeb9e7895fb3b74a","cross_cats_sorted":["cs.HC","physics.data-an"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T07:31:44Z","title_canon_sha256":"78e1cae135e4185ba867359a42d4e2e863d2184fd6bfe9e18c49eebba86d9ea1"},"schema_version":"1.0","source":{"id":"2606.00093","kind":"arxiv","version":1}},"canonical_sha256":"694217726bfcb0e97d4cbf1f1cea38a7f16936ed10ec9c13149bc781c0bfa058","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"694217726bfcb0e97d4cbf1f1cea38a7f16936ed10ec9c13149bc781c0bfa058","first_computed_at":"2026-06-02T00:03:14.787278Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T00:03:14.787278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PymqJeajwk46c0h54wtJ+UIJPVfL8RXsncqZPA1rbS2XoIFJyQFwQbBoWNCBdW59CILo2qQqfXecvl89DlbYAA==","signature_status":"signed_v1","signed_at":"2026-06-02T00:03:14.787695Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00093","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f5f3dc2109ac2f714068fbd6c6c543da28a85227df4ead4249e52917872c46e0","sha256:f9766f5c422b1b3aca5b4a503ea75279b7065818c2fe04d474975004f9f2005a"],"state_sha256":"acf2f8c7a1bd4d0e9c434ed80b81c85f3fd58b7fec3b6f291cb73979dce48715"}