{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:QDIHNUJ3ESG7TOMYROS2KILMJ3","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":"2874148d47759e6dfb5cfac499f2575acb626443cc413bb378500da2148d5ce3","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-06-23T20:02:55Z","title_canon_sha256":"3582e498633daa810fc8a54e6a0d355d3ae133b70c47b2e915ae4efc6e45b1f7"},"schema_version":"1.0","source":{"id":"1806.09029","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.09029","created_at":"2026-07-05T01:07:48Z"},{"alias_kind":"arxiv_version","alias_value":"1806.09029v1","created_at":"2026-07-05T01:07:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.09029","created_at":"2026-07-05T01:07:48Z"},{"alias_kind":"pith_short_12","alias_value":"QDIHNUJ3ESG7","created_at":"2026-07-05T01:07:48Z"},{"alias_kind":"pith_short_16","alias_value":"QDIHNUJ3ESG7TOMY","created_at":"2026-07-05T01:07:48Z"},{"alias_kind":"pith_short_8","alias_value":"QDIHNUJ3","created_at":"2026-07-05T01:07:48Z"}],"graph_snapshots":[{"event_id":"sha256:447455770941a27de6008a96df0dddca93ed9c7e0f819207e277116fa92e34f0","target":"graph","created_at":"2026-07-05T01:07:48Z","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/1806.09029/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To be informative, an evaluation must measure how well systems generalize to realistic unseen data. We identify limitations of and propose improvements to current evaluations of text-to-SQL systems. First, we compare human-generated and automatically generated questions, characterizing properties of queries necessary for real-world applications. To facilitate evaluation on multiple datasets, we release standardized and improved versions of seven existing datasets and one new text-to-SQL dataset. Second, we show that the current division of data into training and test sets measures robustness t","authors_text":"Catherine Finegan-Dollak, Dragomir Radev, Jonathan K. Kummerfeld, Karthik Ramanathan, Li Zhang, Rui Zhang, Sesh Sadasivam","cross_cats":["cs.AI","cs.DB"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-06-23T20:02:55Z","title":"Improving Text-to-SQL Evaluation Methodology"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.09029","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:da33450645814386665004108998ffda03d45fc46e1584e0e75d7c986faf3c60","target":"record","created_at":"2026-07-05T01:07:48Z","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":"2874148d47759e6dfb5cfac499f2575acb626443cc413bb378500da2148d5ce3","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-06-23T20:02:55Z","title_canon_sha256":"3582e498633daa810fc8a54e6a0d355d3ae133b70c47b2e915ae4efc6e45b1f7"},"schema_version":"1.0","source":{"id":"1806.09029","kind":"arxiv","version":1}},"canonical_sha256":"80d076d13b248df9b9988ba5a5216c4ee47ba09b51bd92503f5d55c5bf9c1e54","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"80d076d13b248df9b9988ba5a5216c4ee47ba09b51bd92503f5d55c5bf9c1e54","first_computed_at":"2026-07-05T01:07:48.021031Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:07:48.021031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"a4Pi65jmlBJyLGXezRSLns5Kam8et7yBWZhIMBGVev1cpSoxymFapIgKFDKsAAx5yM8RkbpzAUJD51S1AABnAw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:07:48.021535Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.09029","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:da33450645814386665004108998ffda03d45fc46e1584e0e75d7c986faf3c60","sha256:447455770941a27de6008a96df0dddca93ed9c7e0f819207e277116fa92e34f0"],"state_sha256":"7f96f057f47eed21ad5ff8f525fd47e499bda08f720fa5241b92e98924a47bb8"}