{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AP2TBU3KQLX4IMIIBMUBMH37MZ","short_pith_number":"pith:AP2TBU3K","schema_version":"1.0","canonical_sha256":"03f530d36a82efc431080b28161f7f6655fcea77ad1e2d32189bbd644d22f809","source":{"kind":"arxiv","id":"2606.05169","version":1},"attestation_state":"computed","paper":{"title":"The Evaluation Blind Spot: A Stereological Theory of Benchmark Coverage for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jason Z Wang","submitted_at":"2026-04-15T08:56:58Z","abstract_excerpt":"We give a stereological theory of LLM benchmark coverage. For any suite with effective dimensionality d_eff, the visible Hausdorff distance between two convex capability profiles consistent with the same scores is bounded by epsilon + C R m^(-1/(d_eff-1)), with matching Lipschitz lower bound. Empirically, three independent leaderboards (Open LLM v2, an extended 12-benchmark suite, LiveBench) all have d_eff in [2.86, 4.80] on their competitive frontier; the structural blind spot exceeds the observed runner-up score gap by two orders of magnitude and dominates statistical noise by 52-127x. Under"},"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":"2606.05169","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-15T08:56:58Z","cross_cats_sorted":[],"title_canon_sha256":"3b53d443e2b19729e28685b4a460f54104bcc333137a943f25399aa5a41faa77","abstract_canon_sha256":"adbdb6e42994dec504bc54e0eab254dfb84c043a36fae76fb950269283e9acac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T00:13:47.549857Z","signature_b64":"kLkQZwtlrAVk+C/KzZZ3cIOPSOgkSLNFRAL1cnSVXSo/D0rE9x40ldPfJUJLioZXxKQnHMQ3UvmCjIjV2fc/Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03f530d36a82efc431080b28161f7f6655fcea77ad1e2d32189bbd644d22f809","last_reissued_at":"2026-06-05T00:13:47.549250Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T00:13:47.549250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Evaluation Blind Spot: A Stereological Theory of Benchmark Coverage for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jason Z Wang","submitted_at":"2026-04-15T08:56:58Z","abstract_excerpt":"We give a stereological theory of LLM benchmark coverage. For any suite with effective dimensionality d_eff, the visible Hausdorff distance between two convex capability profiles consistent with the same scores is bounded by epsilon + C R m^(-1/(d_eff-1)), with matching Lipschitz lower bound. Empirically, three independent leaderboards (Open LLM v2, an extended 12-benchmark suite, LiveBench) all have d_eff in [2.86, 4.80] on their competitive frontier; the structural blind spot exceeds the observed runner-up score gap by two orders of magnitude and dominates statistical noise by 52-127x. Under"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05169","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/2606.05169/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":"2606.05169","created_at":"2026-06-05T00:13:47.549325+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05169v1","created_at":"2026-06-05T00:13:47.549325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05169","created_at":"2026-06-05T00:13:47.549325+00:00"},{"alias_kind":"pith_short_12","alias_value":"AP2TBU3KQLX4","created_at":"2026-06-05T00:13:47.549325+00:00"},{"alias_kind":"pith_short_16","alias_value":"AP2TBU3KQLX4IMII","created_at":"2026-06-05T00:13:47.549325+00:00"},{"alias_kind":"pith_short_8","alias_value":"AP2TBU3K","created_at":"2026-06-05T00:13:47.549325+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/AP2TBU3KQLX4IMIIBMUBMH37MZ","json":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ.json","graph_json":"https://pith.science/api/pith-number/AP2TBU3KQLX4IMIIBMUBMH37MZ/graph.json","events_json":"https://pith.science/api/pith-number/AP2TBU3KQLX4IMIIBMUBMH37MZ/events.json","paper":"https://pith.science/paper/AP2TBU3K"},"agent_actions":{"view_html":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ","download_json":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ.json","view_paper":"https://pith.science/paper/AP2TBU3K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05169&json=true","fetch_graph":"https://pith.science/api/pith-number/AP2TBU3KQLX4IMIIBMUBMH37MZ/graph.json","fetch_events":"https://pith.science/api/pith-number/AP2TBU3KQLX4IMIIBMUBMH37MZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ/action/storage_attestation","attest_author":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ/action/author_attestation","sign_citation":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ/action/citation_signature","submit_replication":"https://pith.science/pith/AP2TBU3KQLX4IMIIBMUBMH37MZ/action/replication_record"}},"created_at":"2026-06-05T00:13:47.549325+00:00","updated_at":"2026-06-05T00:13:47.549325+00:00"}