{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EF7H7GEUPPNGXR4MHBUV2K32US","short_pith_number":"pith:EF7H7GEU","schema_version":"1.0","canonical_sha256":"217e7f98947bda6bc78c38695d2b7aa4af6a31e08314e6e607dd43f3128b795e","source":{"kind":"arxiv","id":"2607.06799","version":1},"attestation_state":"computed","paper":{"title":"What Predicts Correctness in Text-to-SQL? A Selective-Prediction Study","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Robert Richardson","submitted_at":"2026-07-07T20:52:58Z","abstract_excerpt":"Evaluating uncertainty in AI-generated SQL queries requires estimating whether a query is correct, where correct means it executes to the same result as a human-written reference. We study which signals predict correctness on hard multi-table text-to-SQL, using AUROC to measure how well each ranks correct queries above incorrect ones. On BIRD and Spider, black-box signals such as string, structural, and execution self-consistency, a schema-relevance score, and query executability all fall between about 0.61 and 0.68 AUROC, with string self-consistency strongest at 0.675; white-box log-probabil"},"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":"2607.06799","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-07T20:52:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6cc2be45e6daa88647f592a93d63f2fc90d93f0c3a25122172e1f8832948ee58","abstract_canon_sha256":"ce84de958cc8adc664efe4fe49b6a0719b4505a9a8c1b923880e97d3e07a8854"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-09T00:19:28.010586Z","signature_b64":"aOyQeTUzvMhU421kvxd9fkwwpFVZZwkVt0UpYf4xljR7OqIhOxJo62WG0WWoS5qOO6gDw/ttfIEwH0UDByPDAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"217e7f98947bda6bc78c38695d2b7aa4af6a31e08314e6e607dd43f3128b795e","last_reissued_at":"2026-07-09T00:19:28.010180Z","signature_status":"signed_v1","first_computed_at":"2026-07-09T00:19:28.010180Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Predicts Correctness in Text-to-SQL? A Selective-Prediction Study","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Robert Richardson","submitted_at":"2026-07-07T20:52:58Z","abstract_excerpt":"Evaluating uncertainty in AI-generated SQL queries requires estimating whether a query is correct, where correct means it executes to the same result as a human-written reference. We study which signals predict correctness on hard multi-table text-to-SQL, using AUROC to measure how well each ranks correct queries above incorrect ones. On BIRD and Spider, black-box signals such as string, structural, and execution self-consistency, a schema-relevance score, and query executability all fall between about 0.61 and 0.68 AUROC, with string self-consistency strongest at 0.675; white-box log-probabil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.06799","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/2607.06799/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":"2607.06799","created_at":"2026-07-09T00:19:28.010233+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.06799v1","created_at":"2026-07-09T00:19:28.010233+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.06799","created_at":"2026-07-09T00:19:28.010233+00:00"},{"alias_kind":"pith_short_12","alias_value":"EF7H7GEUPPNG","created_at":"2026-07-09T00:19:28.010233+00:00"},{"alias_kind":"pith_short_16","alias_value":"EF7H7GEUPPNGXR4M","created_at":"2026-07-09T00:19:28.010233+00:00"},{"alias_kind":"pith_short_8","alias_value":"EF7H7GEU","created_at":"2026-07-09T00:19:28.010233+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.07946","citing_title":"Metadata Reconstruction from Values Alone: Recovering Column Semantics in Undocumented Warehouses","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US","json":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US.json","graph_json":"https://pith.science/api/pith-number/EF7H7GEUPPNGXR4MHBUV2K32US/graph.json","events_json":"https://pith.science/api/pith-number/EF7H7GEUPPNGXR4MHBUV2K32US/events.json","paper":"https://pith.science/paper/EF7H7GEU"},"agent_actions":{"view_html":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US","download_json":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US.json","view_paper":"https://pith.science/paper/EF7H7GEU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.06799&json=true","fetch_graph":"https://pith.science/api/pith-number/EF7H7GEUPPNGXR4MHBUV2K32US/graph.json","fetch_events":"https://pith.science/api/pith-number/EF7H7GEUPPNGXR4MHBUV2K32US/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US/action/storage_attestation","attest_author":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US/action/author_attestation","sign_citation":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US/action/citation_signature","submit_replication":"https://pith.science/pith/EF7H7GEUPPNGXR4MHBUV2K32US/action/replication_record"}},"created_at":"2026-07-09T00:19:28.010233+00:00","updated_at":"2026-07-09T00:19:28.010233+00:00"}