{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UE47YYAL6OFTPJ2DBUISYRO5HK","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":"4ddfc97ec4f37f263ec4a98c5b8003a8316d2c83bcb2438fc5fcfdbd01fba98d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-30T04:38:18Z","title_canon_sha256":"c7146fe3ac3763d080d19234f945dd02fe6d1d265a0c338466d051e5323246e2"},"schema_version":"1.0","source":{"id":"2601.22548","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.22548","created_at":"2026-06-24T00:14:22Z"},{"alias_kind":"arxiv_version","alias_value":"2601.22548v4","created_at":"2026-06-24T00:14:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.22548","created_at":"2026-06-24T00:14:22Z"},{"alias_kind":"pith_short_12","alias_value":"UE47YYAL6OFT","created_at":"2026-06-24T00:14:22Z"},{"alias_kind":"pith_short_16","alias_value":"UE47YYAL6OFTPJ2D","created_at":"2026-06-24T00:14:22Z"},{"alias_kind":"pith_short_8","alias_value":"UE47YYAL","created_at":"2026-06-24T00:14:22Z"}],"graph_snapshots":[{"event_id":"sha256:9ebb0ebb4edcadb435899b1afcebce1458c7b9e7cb922c45993998b635b1e11a","target":"graph","created_at":"2026-06-24T00:14:22Z","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/2601.22548/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent research has shown that large language models (LLMs) favor their own outputs when acting as judges, undermining the integrity of automated post-training and evaluation workflows. However, it is difficult to disentangle which behaviors are explained by narcissism versus experimental confounds. Specifically, LLM evaluators may deliver self-preferring verdicts when comparing responses to questions they fail on; these verdicts may not depend on the identity of the author, but on evaluator quality. We correct this by directly comparing the judge's voting distribution in cases where it evalua","authors_text":"Dani Roytburg, Jou Barzdukas, Mackenzie Puig-Hall, Matthew Bozoukov, Matthew Nguyen, Narmeen Oozeer","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-30T04:38:18Z","title":"Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.22548","kind":"arxiv","version":4},"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:edf536080fc904d1bed45e5a0146c5616584be87e0792bdb6a3fb508231ae541","target":"record","created_at":"2026-06-24T00:14:22Z","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":"4ddfc97ec4f37f263ec4a98c5b8003a8316d2c83bcb2438fc5fcfdbd01fba98d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-30T04:38:18Z","title_canon_sha256":"c7146fe3ac3763d080d19234f945dd02fe6d1d265a0c338466d051e5323246e2"},"schema_version":"1.0","source":{"id":"2601.22548","kind":"arxiv","version":4}},"canonical_sha256":"a139fc600bf38b37a7430d112c45dd3ab825151ed0667da82333433d6f9ce2df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a139fc600bf38b37a7430d112c45dd3ab825151ed0667da82333433d6f9ce2df","first_computed_at":"2026-06-24T00:14:22.353959Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-24T00:14:22.353959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cJQnLCoBqJgUVhz+1ZZjs1usP31YlMGGqiP/87cstLXZkTATfKNzOcTIv63/mN2ZbmiYFT78X+uxOarZCE4cDA==","signature_status":"signed_v1","signed_at":"2026-06-24T00:14:22.354394Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.22548","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:edf536080fc904d1bed45e5a0146c5616584be87e0792bdb6a3fb508231ae541","sha256:9ebb0ebb4edcadb435899b1afcebce1458c7b9e7cb922c45993998b635b1e11a"],"state_sha256":"0089bb40e7f73a94e580a90aa59639cab6a933cf0976abc3a3e622f30a4439f1"}