{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TPRWGTWN5VGQXWI5QWKVTDFCYX","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":"051e085c11cd1181c13835a915fdbc81ae7ccb28b938305793c4598545f5bf1e","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-29T12:33:06Z","title_canon_sha256":"ac39e824a6866086a1c7e7c6a5f9cc99f056ed8e5ed7002f4ddf63ff9ca30589"},"schema_version":"1.0","source":{"id":"2606.30219","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.30219","created_at":"2026-06-30T02:17:54Z"},{"alias_kind":"arxiv_version","alias_value":"2606.30219v1","created_at":"2026-06-30T02:17:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30219","created_at":"2026-06-30T02:17:54Z"},{"alias_kind":"pith_short_12","alias_value":"TPRWGTWN5VGQ","created_at":"2026-06-30T02:17:54Z"},{"alias_kind":"pith_short_16","alias_value":"TPRWGTWN5VGQXWI5","created_at":"2026-06-30T02:17:54Z"},{"alias_kind":"pith_short_8","alias_value":"TPRWGTWN","created_at":"2026-06-30T02:17:54Z"}],"graph_snapshots":[{"event_id":"sha256:b6aceb0190397240839fa8132d5d6d9199c5c8667a969a66335bc3a8e870ba47","target":"graph","created_at":"2026-06-30T02:17:54Z","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/2606.30219/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM evaluation and AI safety face a shared measurement problem: benchmark scores, reward-model signals, and reported safety metrics can improve while the latent properties they are meant to represent remain difficult to verify. This paper combines a hybrid survey - a systematic search paired with narrative synthesis and separately tracked grey evidence - with a conceptual framework and a structured ten-model audit. The synthesis spans eight evidence streams: benchmark validity, dynamic evaluation, LLM-as-judge reliability, safety evaluation, jailbreak/refusal robustness, reward hacking, mechan","authors_text":"Bu\\u{g}ra Alperen Ulu{\\i}rmak, Rifat Kurban","cross_cats":["cs.CL","cs.LG","cs.SE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-29T12:33:06Z","title":"EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30219","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":"sha256:39f9ccea35d47acb84a72429e176325742c7cfe08400b81505aed136864d1169","target":"record","created_at":"2026-06-30T02:17:54Z","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":"051e085c11cd1181c13835a915fdbc81ae7ccb28b938305793c4598545f5bf1e","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-29T12:33:06Z","title_canon_sha256":"ac39e824a6866086a1c7e7c6a5f9cc99f056ed8e5ed7002f4ddf63ff9ca30589"},"schema_version":"1.0","source":{"id":"2606.30219","kind":"arxiv","version":1}},"canonical_sha256":"9be3634ecded4d0bd91d8595598ca2c5c5cd6a5b3c3f9e5ba1fc7d4379336d13","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9be3634ecded4d0bd91d8595598ca2c5c5cd6a5b3c3f9e5ba1fc7d4379336d13","first_computed_at":"2026-06-30T02:17:54.897279Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:17:54.897279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"U4VTBnASHo+/XEpY5tkdpUDVOTz9Ui3EMAdicHbj3HgPy2tk2zVXoqOLOidH4CW8/vtbI6/R6lbnf9IvISHnBg==","signature_status":"signed_v1","signed_at":"2026-06-30T02:17:54.898076Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.30219","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:39f9ccea35d47acb84a72429e176325742c7cfe08400b81505aed136864d1169","sha256:b6aceb0190397240839fa8132d5d6d9199c5c8667a969a66335bc3a8e870ba47"],"state_sha256":"d4176d5b3098475127053d2e242309f4dc0605c028fb353c32ce1e587c919eb3"}