{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TPRWGTWN5VGQXWI5QWKVTDFCYX","short_pith_number":"pith:TPRWGTWN","schema_version":"1.0","canonical_sha256":"9be3634ecded4d0bd91d8595598ca2c5c5cd6a5b3c3f9e5ba1fc7d4379336d13","source":{"kind":"arxiv","id":"2606.30219","version":1},"attestation_state":"computed","paper":{"title":"EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SE"],"primary_cat":"cs.AI","authors_text":"Bu\\u{g}ra Alperen Ulu{\\i}rmak, Rifat Kurban","submitted_at":"2026-06-29T12:33:06Z","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"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.30219","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-29T12:33:06Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"title_canon_sha256":"ac39e824a6866086a1c7e7c6a5f9cc99f056ed8e5ed7002f4ddf63ff9ca30589","abstract_canon_sha256":"051e085c11cd1181c13835a915fdbc81ae7ccb28b938305793c4598545f5bf1e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:54.898076Z","signature_b64":"U4VTBnASHo+/XEpY5tkdpUDVOTz9Ui3EMAdicHbj3HgPy2tk2zVXoqOLOidH4CW8/vtbI6/R6lbnf9IvISHnBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9be3634ecded4d0bd91d8595598ca2c5c5cd6a5b3c3f9e5ba1fc7d4379336d13","last_reissued_at":"2026-06-30T02:17:54.897279Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:54.897279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SE"],"primary_cat":"cs.AI","authors_text":"Bu\\u{g}ra Alperen Ulu{\\i}rmak, Rifat Kurban","submitted_at":"2026-06-29T12:33:06Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30219","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.30219/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.30219","created_at":"2026-06-30T02:17:54.897391+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.30219v1","created_at":"2026-06-30T02:17:54.897391+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30219","created_at":"2026-06-30T02:17:54.897391+00:00"},{"alias_kind":"pith_short_12","alias_value":"TPRWGTWN5VGQ","created_at":"2026-06-30T02:17:54.897391+00:00"},{"alias_kind":"pith_short_16","alias_value":"TPRWGTWN5VGQXWI5","created_at":"2026-06-30T02:17:54.897391+00:00"},{"alias_kind":"pith_short_8","alias_value":"TPRWGTWN","created_at":"2026-06-30T02:17:54.897391+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX","json":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX.json","graph_json":"https://pith.science/api/pith-number/TPRWGTWN5VGQXWI5QWKVTDFCYX/graph.json","events_json":"https://pith.science/api/pith-number/TPRWGTWN5VGQXWI5QWKVTDFCYX/events.json","paper":"https://pith.science/paper/TPRWGTWN"},"agent_actions":{"view_html":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX","download_json":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX.json","view_paper":"https://pith.science/paper/TPRWGTWN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.30219&json=true","fetch_graph":"https://pith.science/api/pith-number/TPRWGTWN5VGQXWI5QWKVTDFCYX/graph.json","fetch_events":"https://pith.science/api/pith-number/TPRWGTWN5VGQXWI5QWKVTDFCYX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX/action/storage_attestation","attest_author":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX/action/author_attestation","sign_citation":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX/action/citation_signature","submit_replication":"https://pith.science/pith/TPRWGTWN5VGQXWI5QWKVTDFCYX/action/replication_record"}},"created_at":"2026-06-30T02:17:54.897391+00:00","updated_at":"2026-06-30T02:17:54.897391+00:00"}