{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6RVQDMX6LNSGEFKI7KQIP2WIML","merge_version":"pith-open-graph-merge-v1","event_count":15,"valid_event_count":15,"invalid_event_count":0,"equivocation_count":1,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9f25f9b88c9998596b9668523f25a8fe125a8ae9668c6909728524017f3b6cf8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-21T16:06:34Z","title_canon_sha256":"a91cdf966f2e59634533f7af937414a541324ce8776e86b2f86bcce8f605d10a"},"schema_version":"1.0","source":{"id":"2605.22664","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22664","created_at":"2026-05-22T02:04:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22664v1","created_at":"2026-05-22T02:04:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22664","created_at":"2026-05-22T02:04:49Z"},{"alias_kind":"pith_short_12","alias_value":"6RVQDMX6LNSG","created_at":"2026-05-22T02:04:49Z"},{"alias_kind":"pith_short_16","alias_value":"6RVQDMX6LNSGEFKI","created_at":"2026-05-22T02:04:49Z"},{"alias_kind":"pith_short_8","alias_value":"6RVQDMX6","created_at":"2026-05-22T02:04:49Z"}],"graph_snapshots":[{"event_id":"sha256:2df9b58c43f83b3b3912b286c9716adffde140f57b1094b8db2eb85bcd82ba29","target":"graph","created_at":"2026-05-22T02:04:49Z","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/2605.22664/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM agents are increasingly expected to carry out end-to-end workflows, producing complete artifacts from high-level user instructions. To meet enterprise needs, frontier AI labs have developed agents that can construct entire spreadsheets from scratch. This is especially relevant in finance, where core workflows such as financial modeling, forecasting, and scenario analysis are commonly conducted through spreadsheets. Yet, existing spreadsheet benchmarks do not measure this advanced capability, focusing instead on question-answering or single-formula edits. To address this gap, we provide one","authors_text":"Adam Shen, Ali Bauyrzhan, Daniel Guetta, Haoyang Liu, Harshith Srinivas Gear, Hongseok Namkoong, Joshua Fan, Julian Poeltl, Siri Du, Thomson Yen, Yili Liu, Yilin Meng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-21T16:06:34Z","title":"WorkstreamBench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22664","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:dc97d7fd2732410f37d08b1145c5b9cb526d87eb4e7712a8435ad9e691a6b7a0","target":"record","created_at":"2026-05-22T02:04:49Z","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":"9f25f9b88c9998596b9668523f25a8fe125a8ae9668c6909728524017f3b6cf8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-21T16:06:34Z","title_canon_sha256":"a91cdf966f2e59634533f7af937414a541324ce8776e86b2f86bcce8f605d10a"},"schema_version":"1.0","source":{"id":"2605.22664","kind":"arxiv","version":1}},"canonical_sha256":"f46b01b2fe5b64621548faa087eac862cb92023774177149100d3eafe5359071","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f46b01b2fe5b64621548faa087eac862cb92023774177149100d3eafe5359071","first_computed_at":"2026-05-22T02:04:49.132723Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T02:04:49.132723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8zO8uJjotOK8lgf5kQAuliGl79gH/EICPuNpWOspGdiYqS7HHW1A9jM11onNtS1ip5R2npKImaaOP55M38IICQ==","signature_status":"signed_v1","signed_at":"2026-05-22T02:04:49.133588Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.22664","source_kind":"arxiv","source_version":1}}},"equivocations":[{"signer_id":"pith.science","event_type":"integrity_finding","target":"integrity","event_ids":["sha256:054bc57dfa335feec76da4e8698569c2b490021df44513fcab5a86c0c7a58db7","sha256:12f53f3ff2a8361e45439d7cf8655e8330c284f639c69f6878f91fcf646f61c6","sha256:5725c8019f3e0a011dbf7d83ae67963d0bdcd86916b3e55ee722bc28aed1d828","sha256:581639335e9bc1d46b608de1abffc85ea3cbcb6ca8a13bfd200414a5e67c0b57","sha256:6eea03368d916a58d8250b321dbbd14f2d12eb5b9f83f1203154be68143001f2","sha256:7398f97afd26645f6d68ba2f5e04b2524a7b0bd65278433d0ab8c4367da9959e","sha256:73b403912c46cd34d112b0c09a8546541f5b8e184641452093fc15ad9fb60ce0","sha256:7a00a0498e554025f867acb2c08723d8a402ea4df7bf7a70378f93a4ba4e57f6","sha256:db3fda94b806e029e714fb4a19870d6daa055de3aad897356399f99283bafe95","sha256:f0bc069c849f61006ff58822430a444d486104fbff7057c498b3c3915fee9d59","sha256:f7d5693a1d064827d651b11ccc6f5fe60f2d631e6904211f6234492fa1f843f9","sha256:fa2d89f5a51bd622fa59858fa563dd105e1b5fd6bc47ea0afad99f193ab10db1","sha256:fd3b527d0f41711bbf568ca416a5b40846fd03e95adebca5794e24c1582987ed"]}],"invalid_events":[],"applied_event_ids":["sha256:dc97d7fd2732410f37d08b1145c5b9cb526d87eb4e7712a8435ad9e691a6b7a0","sha256:2df9b58c43f83b3b3912b286c9716adffde140f57b1094b8db2eb85bcd82ba29"],"state_sha256":"9b6e55e7d1b824c24cad8e9c95c50fd87d4db861a8fedc39c7631123fffef4c7"}