{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:F6XERQ72HKF42FGFUJLZRJWDQ4","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":"5b7af69b25227fe8e60ac942c298b064bc5edf0823a8916909ac4fe79e0fb3e3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T14:50:08Z","title_canon_sha256":"31121cc6c946e328f51c177179993d070977c03e0f06b4e866b326cd6c7f7bb2"},"schema_version":"1.0","source":{"id":"2605.31381","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.31381","created_at":"2026-06-01T02:04:02Z"},{"alias_kind":"arxiv_version","alias_value":"2605.31381v1","created_at":"2026-06-01T02:04:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.31381","created_at":"2026-06-01T02:04:02Z"},{"alias_kind":"pith_short_12","alias_value":"F6XERQ72HKF4","created_at":"2026-06-01T02:04:02Z"},{"alias_kind":"pith_short_16","alias_value":"F6XERQ72HKF42FGF","created_at":"2026-06-01T02:04:02Z"},{"alias_kind":"pith_short_8","alias_value":"F6XERQ72","created_at":"2026-06-01T02:04:02Z"}],"graph_snapshots":[{"event_id":"sha256:c06be401b62dd8050143e1496a29bd2aee4a89c3a3de380a1a614b3339fb2a9a","target":"graph","created_at":"2026-06-01T02:04:02Z","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/2605.31381/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We evaluate the consistency of automated judges in conducting a multi-dimensional safety evaluation in a reference-free setup. Our results indicate that Large Language Models are unreliable judges in identifying safety issues related to machine-generated advice in regulated domains such as finance, although they are more reliable at identifying more overt forms of unsafe/harmful content such as violence. The degree of inconsistency in a model's judgments can vary significantly by the chosen safety criteria and can be impacted by the language of the content and its linguistic style as well. Fin","authors_text":"Akriti Vij, Isar Nejadgholi, Krishnapriya Vishnubhotla, Soumya Vajjala","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T14:50:08Z","title":"LLM Judges Inconsistently Disagree Across Safety Criteria and Harm Categories"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.31381","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":"sha256:38f522cc7aad21b5928ce4a216ea9c6782f0bfb2e801003611ffbc2332059e72","target":"record","created_at":"2026-06-01T02:04:02Z","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":"5b7af69b25227fe8e60ac942c298b064bc5edf0823a8916909ac4fe79e0fb3e3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T14:50:08Z","title_canon_sha256":"31121cc6c946e328f51c177179993d070977c03e0f06b4e866b326cd6c7f7bb2"},"schema_version":"1.0","source":{"id":"2605.31381","kind":"arxiv","version":1}},"canonical_sha256":"2fae48c3fa3a8bcd14c5a25798a6c3872e7bbf457c9d3b9a95136d66135fe3ae","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2fae48c3fa3a8bcd14c5a25798a6c3872e7bbf457c9d3b9a95136d66135fe3ae","first_computed_at":"2026-06-01T02:04:02.139551Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-01T02:04:02.139551Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9KeZJdlSuBA7O9UxfdK2rs5hHND+8jXDZLLiCE1R7654RfuNF21wGfV3kgqn3WXqclRZWehekTPoHHq2t7GGCA==","signature_status":"signed_v1","signed_at":"2026-06-01T02:04:02.140670Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.31381","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:38f522cc7aad21b5928ce4a216ea9c6782f0bfb2e801003611ffbc2332059e72","sha256:c06be401b62dd8050143e1496a29bd2aee4a89c3a3de380a1a614b3339fb2a9a"],"state_sha256":"9c9d3ef9f96fc738e321bed84cb41966138365e82fd594b2bc48ffabdc2c3d2d"}