{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:J74XWQNW25KC7WFSH34FTHUX4M","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":"59c4f2c1fdbdf531480d34a04c4d4bc697531d28e897f1fb46f73e0dbea1086a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T17:11:56Z","title_canon_sha256":"e8f0bcc884f9a14e497563a9201c1e3139d3477727d6241509a9b9c20ddde375"},"schema_version":"1.0","source":{"id":"2606.03918","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.03918","created_at":"2026-06-03T02:06:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.03918v1","created_at":"2026-06-03T02:06:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03918","created_at":"2026-06-03T02:06:06Z"},{"alias_kind":"pith_short_12","alias_value":"J74XWQNW25KC","created_at":"2026-06-03T02:06:06Z"},{"alias_kind":"pith_short_16","alias_value":"J74XWQNW25KC7WFS","created_at":"2026-06-03T02:06:06Z"},{"alias_kind":"pith_short_8","alias_value":"J74XWQNW","created_at":"2026-06-03T02:06:06Z"}],"graph_snapshots":[{"event_id":"sha256:6d4ee80f1ed81b045eee4b96f42ed3f21a664810f9f054758a2a1207b027ac34","target":"graph","created_at":"2026-06-03T02:06:06Z","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/2606.03918/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AI agents can increasingly handle the mechanical tasks of financial analysis: retrieving documents, calculating formulas, updating spreadsheets. The harder, more valuable challenge is reasoning through the open-ended questions that define expert Analyst work. Existing benchmarks do not capture this class of problem, and those that attempt to evaluate open-ended reasoning rely on model-judged outputs that introduce noise and circularity. We present Hedge-Bench 1.0: a benchmark of 102 actual, on-the-job tasks grounded in the explicit reasoning traces of professional hedge fund analysts working w","authors_text":"Alice Lu, Andy Lyu, Eric Cho, Shawn Huang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T17:11:56Z","title":"Hedge-Bench: Benchmarking Agents on Hard, Realistic Tasks Pertaining to Financial Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03918","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:b473b21455861c2af91659b87913ee1066581cd20d71cbc55195be38ee1e6bbc","target":"record","created_at":"2026-06-03T02:06:06Z","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":"59c4f2c1fdbdf531480d34a04c4d4bc697531d28e897f1fb46f73e0dbea1086a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T17:11:56Z","title_canon_sha256":"e8f0bcc884f9a14e497563a9201c1e3139d3477727d6241509a9b9c20ddde375"},"schema_version":"1.0","source":{"id":"2606.03918","kind":"arxiv","version":1}},"canonical_sha256":"4ff97b41b6d7542fd8b23ef8599e97e32372aa688be6ed487ad1676da78f22f9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4ff97b41b6d7542fd8b23ef8599e97e32372aa688be6ed487ad1676da78f22f9","first_computed_at":"2026-06-03T02:06:06.858073Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-03T02:06:06.858073Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MvJoTNPHPR6vCnwzB5YlKjo0ABORYzUVsuazfOFP7zg21ftQpWVNRFUngs9BPbdEeGH36WwGWmsx82hEBPDsBg==","signature_status":"signed_v1","signed_at":"2026-06-03T02:06:06.858491Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.03918","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b473b21455861c2af91659b87913ee1066581cd20d71cbc55195be38ee1e6bbc","sha256:6d4ee80f1ed81b045eee4b96f42ed3f21a664810f9f054758a2a1207b027ac34"],"state_sha256":"4f7b31b6fb8adb814ec1b83dd303ce5165f1c5bcc0e216d408abbbf05bc8bdbd"}