{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NZO2BNT2IDSH6JS24KZ3MKD5OP","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":"519d19621b099075983c6c81b595d1e078d323cdd1a0f8a08ff2eec451312968","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-11T16:16:11Z","title_canon_sha256":"fae189561a94b659bcbff9be9f01767c36220ef08663f16d6d8dac4c3f526069"},"schema_version":"1.0","source":{"id":"2311.06602","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.06602","created_at":"2026-07-05T07:55:13Z"},{"alias_kind":"arxiv_version","alias_value":"2311.06602v2","created_at":"2026-07-05T07:55:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.06602","created_at":"2026-07-05T07:55:13Z"},{"alias_kind":"pith_short_12","alias_value":"NZO2BNT2IDSH","created_at":"2026-07-05T07:55:13Z"},{"alias_kind":"pith_short_16","alias_value":"NZO2BNT2IDSH6JS2","created_at":"2026-07-05T07:55:13Z"},{"alias_kind":"pith_short_8","alias_value":"NZO2BNT2","created_at":"2026-07-05T07:55:13Z"}],"graph_snapshots":[{"event_id":"sha256:0f959d18c18c13290a8f6a40c18fdd06f083de56d1a1666b87b013b34b136d1d","target":"graph","created_at":"2026-07-05T07:55:13Z","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/2311.06602/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Answering questions within business and finance requires reasoning, precision, and a wide-breadth of technical knowledge. Together, these requirements make this domain difficult for large language models (LLMs). We introduce BizBench, a benchmark for evaluating models' ability to reason about realistic financial problems. BizBench comprises eight quantitative reasoning tasks, focusing on question-answering (QA) over financial data via program synthesis. We include three financially-themed code-generation tasks from newly collected and augmented QA data. Additionally, we isolate the reasoning c","authors_text":"Charles Lovering, Chris Tanner, Michael Krumdick, Rik Koncel-Kedziorski, Varshini Reddy, Viet Lai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-11T16:16:11Z","title":"BizBench: A Quantitative Reasoning Benchmark for Business and Finance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.06602","kind":"arxiv","version":2},"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":"sha256:9d337f888eb4f1b9bcd4fc914d9f3658988a9cbb1369eb14317b100830717655","target":"record","created_at":"2026-07-05T07:55:13Z","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":"519d19621b099075983c6c81b595d1e078d323cdd1a0f8a08ff2eec451312968","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-11T16:16:11Z","title_canon_sha256":"fae189561a94b659bcbff9be9f01767c36220ef08663f16d6d8dac4c3f526069"},"schema_version":"1.0","source":{"id":"2311.06602","kind":"arxiv","version":2}},"canonical_sha256":"6e5da0b67a40e47f265ae2b3b6287d73d7850f6e71d8209570a8dee31d21a955","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6e5da0b67a40e47f265ae2b3b6287d73d7850f6e71d8209570a8dee31d21a955","first_computed_at":"2026-07-05T07:55:13.681555Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:55:13.681555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XDoW+Kob6921Dv50fRCVrmshKpG4UiZcP0kl8Kn/7HwxbBk8xbZkN8fKyj++L9sgV9jTZWsnMl7P/eAzna/eCA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:55:13.682137Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.06602","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9d337f888eb4f1b9bcd4fc914d9f3658988a9cbb1369eb14317b100830717655","sha256:0f959d18c18c13290a8f6a40c18fdd06f083de56d1a1666b87b013b34b136d1d"],"state_sha256":"e9bf45db4445ecac13ec54a1c947347af6410e1b8bdbff60a049ae53763e379c"}