{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PPJGS2D3HRN5L2LXZXRSOKIZ7M","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":"6e5b04d502f6061224d4cb37493344abdb44a037e5ab7152deb57fc49b1c5cb5","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T08:14:24Z","title_canon_sha256":"1f43d154f6b055e0361ba77f63f078b0efbc8fb10b4ea70d71874be3a192d77e"},"schema_version":"1.0","source":{"id":"2608.09254","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.09254","created_at":"2026-08-11T02:22:11Z"},{"alias_kind":"arxiv_version","alias_value":"2608.09254v1","created_at":"2026-08-11T02:22:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09254","created_at":"2026-08-11T02:22:11Z"},{"alias_kind":"pith_short_12","alias_value":"PPJGS2D3HRN5","created_at":"2026-08-11T02:22:11Z"},{"alias_kind":"pith_short_16","alias_value":"PPJGS2D3HRN5L2LX","created_at":"2026-08-11T02:22:11Z"},{"alias_kind":"pith_short_8","alias_value":"PPJGS2D3","created_at":"2026-08-11T02:22:11Z"}],"graph_snapshots":[{"event_id":"sha256:363bafcb3e24f0fe769481994878625a33d523e8593adf6abbbc88e03ed56f02","target":"graph","created_at":"2026-08-11T02:22:11Z","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/2608.09254/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM analytics agents are evaluated on SQL syntax accuracy, but production failures look different: questions with two valid business definitions, questions the warehouse cannot answer, deprecated columns after a schema change, and queries that execute successfully while returning the wrong business number. No execution-match metric can score them. This paper introduces WarehouseReliabilityBench, 400 frozen tasks over two synthetic warehouses in which roughly half the correct responses are a clarification, an abstention or a refusal, with pinned denominators and a pre-registered paired bootstra","authors_text":"Morris Lee","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T08:14:24Z","title":"Business Truth, not SQL Accuracy: A Rule-Gated 7B Analytics Agent Outperforms a Direct-Prompted 32B Baseline"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09254","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":"sha256:1c7ad89ae0e397d430cf5dedf15950845a4b6372f50865365e036457f0489ab6","target":"record","created_at":"2026-08-11T02:22:11Z","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":"6e5b04d502f6061224d4cb37493344abdb44a037e5ab7152deb57fc49b1c5cb5","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T08:14:24Z","title_canon_sha256":"1f43d154f6b055e0361ba77f63f078b0efbc8fb10b4ea70d71874be3a192d77e"},"schema_version":"1.0","source":{"id":"2608.09254","kind":"arxiv","version":1}},"canonical_sha256":"7bd269687b3c5bd5e977cde3272919fb2cb22ac215ecfea37495cb68b490f7f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7bd269687b3c5bd5e977cde3272919fb2cb22ac215ecfea37495cb68b490f7f2","first_computed_at":"2026-08-11T02:22:11.266585Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-11T02:22:11.266585Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7awgbuSwaNemtbmo7m+U93YVI7bsNpC4wqWnoXzbGqwp0Tq8LHISi2Xp9FiRTt/XFr85ucdLbT+QrEkw58+pCA==","signature_status":"signed_v1","signed_at":"2026-08-11T02:22:11.268039Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.09254","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1c7ad89ae0e397d430cf5dedf15950845a4b6372f50865365e036457f0489ab6","sha256:363bafcb3e24f0fe769481994878625a33d523e8593adf6abbbc88e03ed56f02"],"state_sha256":"ddebc251d27ef4b7cd47fc6f1e6360e1ca1ace81348e2e2773e09bcb0bc19427"}