{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FARGNDAPJZXXM2K3PX4KVHZZAW","short_pith_number":"pith:FARGNDAP","schema_version":"1.0","canonical_sha256":"2822668c0f4e6f76695b7df8aa9f3905a88ce96d8fcec6956a125bd8a48bc3e5","source":{"kind":"arxiv","id":"2606.03829","version":1},"attestation_state":"computed","paper":{"title":"BigFinanceBench: A Workflow-Grounded Benchmark for Financial-Research Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alex Wang, Chase Blagden, Eric Xu, Georg Meinhardt, Jacob Katz, Joseph H. Kim, Pratyush K. Chaudhary","submitted_at":"2026-06-02T16:12:34Z","abstract_excerpt":"Financial-research answers are decision-relevant only when another analyst can audit how they were produced: which source was chosen, which period and accounting definition were used, which assumptions were made, and how the calculation was performed. Existing finance benchmarks largely evaluate isolated subskills or final answers, leaving the auditable derivation itself under-measured. We introduce BigFinanceBench, a 928-item expert-authored benchmark of open-ended financial-research tasks in which each item pairs a ground-truth reference answer with a point-weighted rubric that decomposes th"},"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":"2606.03829","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T16:12:34Z","cross_cats_sorted":[],"title_canon_sha256":"e0d1b927609430edf6f75d3a40db5c9a614d5318e52d340734ca06d51fa23c81","abstract_canon_sha256":"2bb1e1b33528359adc6319c35cd38fe7d5063569f2a5e0cfc50d14c36801817b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:06:03.538372Z","signature_b64":"Ng1cWTFfCsstQqqqjrcF6SZ8q9vRVot3mOs4oxo0ILnsBYdG/IZha3/Vk1buOKyvueVFZ+N0u1Yimo02OVM5Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2822668c0f4e6f76695b7df8aa9f3905a88ce96d8fcec6956a125bd8a48bc3e5","last_reissued_at":"2026-06-03T02:06:03.537992Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:06:03.537992Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BigFinanceBench: A Workflow-Grounded Benchmark for Financial-Research Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alex Wang, Chase Blagden, Eric Xu, Georg Meinhardt, Jacob Katz, Joseph H. Kim, Pratyush K. Chaudhary","submitted_at":"2026-06-02T16:12:34Z","abstract_excerpt":"Financial-research answers are decision-relevant only when another analyst can audit how they were produced: which source was chosen, which period and accounting definition were used, which assumptions were made, and how the calculation was performed. Existing finance benchmarks largely evaluate isolated subskills or final answers, leaving the auditable derivation itself under-measured. We introduce BigFinanceBench, a 928-item expert-authored benchmark of open-ended financial-research tasks in which each item pairs a ground-truth reference answer with a point-weighted rubric that decomposes th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03829","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/2606.03829/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":"2606.03829","created_at":"2026-06-03T02:06:03.538052+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03829v1","created_at":"2026-06-03T02:06:03.538052+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03829","created_at":"2026-06-03T02:06:03.538052+00:00"},{"alias_kind":"pith_short_12","alias_value":"FARGNDAPJZXX","created_at":"2026-06-03T02:06:03.538052+00:00"},{"alias_kind":"pith_short_16","alias_value":"FARGNDAPJZXXM2K3","created_at":"2026-06-03T02:06:03.538052+00:00"},{"alias_kind":"pith_short_8","alias_value":"FARGNDAP","created_at":"2026-06-03T02:06:03.538052+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/FARGNDAPJZXXM2K3PX4KVHZZAW","json":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW.json","graph_json":"https://pith.science/api/pith-number/FARGNDAPJZXXM2K3PX4KVHZZAW/graph.json","events_json":"https://pith.science/api/pith-number/FARGNDAPJZXXM2K3PX4KVHZZAW/events.json","paper":"https://pith.science/paper/FARGNDAP"},"agent_actions":{"view_html":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW","download_json":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW.json","view_paper":"https://pith.science/paper/FARGNDAP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03829&json=true","fetch_graph":"https://pith.science/api/pith-number/FARGNDAPJZXXM2K3PX4KVHZZAW/graph.json","fetch_events":"https://pith.science/api/pith-number/FARGNDAPJZXXM2K3PX4KVHZZAW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW/action/storage_attestation","attest_author":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW/action/author_attestation","sign_citation":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW/action/citation_signature","submit_replication":"https://pith.science/pith/FARGNDAPJZXXM2K3PX4KVHZZAW/action/replication_record"}},"created_at":"2026-06-03T02:06:03.538052+00:00","updated_at":"2026-06-03T02:06:03.538052+00:00"}