{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GO5ZSFIRZSDLVGN6DWI3QKQJRB","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":"b1a360f761afc1361018f469886193f25a7ec3dd347d343bb061f60339a7e263","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-01T11:42:37Z","title_canon_sha256":"40c129b9e86e49469ef2a5ed7e0ca6dc73732dd31092c191d1f26bfce0de2530"},"schema_version":"1.0","source":{"id":"2606.02109","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.02109","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.02109v1","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02109","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_12","alias_value":"GO5ZSFIRZSDL","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_16","alias_value":"GO5ZSFIRZSDLVGN6","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_8","alias_value":"GO5ZSFIR","created_at":"2026-06-02T02:05:06Z"}],"graph_snapshots":[{"event_id":"sha256:d0d233a173ee415616151986019219e41a09aeccb9029f3854e32f8981b8ee82","target":"graph","created_at":"2026-06-02T02:05:06Z","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/2606.02109/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Enterprise AI systems that translate natural language into SQL queries and orchestrate multi-step agentic reasoning pipelines require evaluation approaches fundamentally different from academic benchmarks. Spider and BIRD established execution-accuracy protocols; G-Eval and RAGAS advanced LLM-based assessment; and recent work such as Spider 2.0, BEAVER, and BIRD-Interact has begun to address enterprise and agentic dimensions. No single framework unifies text-to-SQL assessment with agentic behavior evaluation into a production-grade pipeline calibrated against human expert judgment.\n  We presen","authors_text":"Abhishek Sharma, Dorina Strori, Nathan Miller, Shannon Serrao, Soumitra Chatterjee","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-01T11:42:37Z","title":"BADGER: Bridging Agentic and Deterministic Evaluation for Generative Enterprise Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02109","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:b6f3dc59d80b1b9cdeb091bb26ca1e57c7270b2d9ffd11ceebe56969964af97e","target":"record","created_at":"2026-06-02T02:05:06Z","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":"b1a360f761afc1361018f469886193f25a7ec3dd347d343bb061f60339a7e263","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-01T11:42:37Z","title_canon_sha256":"40c129b9e86e49469ef2a5ed7e0ca6dc73732dd31092c191d1f26bfce0de2530"},"schema_version":"1.0","source":{"id":"2606.02109","kind":"arxiv","version":1}},"canonical_sha256":"33bb991511cc86ba99be1d91b82a09886eaef76c21eea1082b055cf0d037b360","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"33bb991511cc86ba99be1d91b82a09886eaef76c21eea1082b055cf0d037b360","first_computed_at":"2026-06-02T02:05:06.709732Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:05:06.709732Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jhM+dpedTcnykKyV4eFbAyTVMCnj7dKggdkro/hPj7LAH4vIace2ap5gkFk2Z5itaXfzR23VxQTqgv0na0S8Dw==","signature_status":"signed_v1","signed_at":"2026-06-02T02:05:06.710124Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.02109","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6f3dc59d80b1b9cdeb091bb26ca1e57c7270b2d9ffd11ceebe56969964af97e","sha256:d0d233a173ee415616151986019219e41a09aeccb9029f3854e32f8981b8ee82"],"state_sha256":"6fcc3cd60bfcfdbdfe2629fc76cc512b17cd9b74a14a8cc92728e59e42d2be70"}