{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IBY5SVFD35JHUSKGEQMXTS3SWK","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":"f3ad9c05b1378f9c0a9c12e54996e14b551711e8b05d578e8a973477e74d519a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T09:02:04Z","title_canon_sha256":"25258c069a575d203368a6a58c06b00c2a7951010f3e17ca694a837765b7d57a"},"schema_version":"1.0","source":{"id":"2506.18421","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18421","created_at":"2026-07-05T11:36:33Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18421v2","created_at":"2026-07-05T11:36:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18421","created_at":"2026-07-05T11:36:33Z"},{"alias_kind":"pith_short_12","alias_value":"IBY5SVFD35JH","created_at":"2026-07-05T11:36:33Z"},{"alias_kind":"pith_short_16","alias_value":"IBY5SVFD35JHUSKG","created_at":"2026-07-05T11:36:33Z"},{"alias_kind":"pith_short_8","alias_value":"IBY5SVFD","created_at":"2026-07-05T11:36:33Z"}],"graph_snapshots":[{"event_id":"sha256:ad6e9217b048e9e71de76e958a95ac49a96792acb80f61e8bcf9c44d5ccc3700","target":"graph","created_at":"2026-07-05T11:36:33Z","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/2506.18421/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The majority of data in businesses and industries is stored in tables, databases, and data warehouses. Reasoning with table-structured data poses significant challenges for large language models (LLMs) due to its hidden semantics, inherent complexity, and structured nature. One of these challenges is lacking an effective evaluation benchmark fairly reflecting the performances of LLMs on broad table reasoning abilities. In this paper, we fill in this gap, presenting a comprehensive table reasoning evolution benchmark, TReB, which measures both shallow table understanding abilities and deep tabl","authors_text":"Boshen Shi, Ce Chi, Ce Li, Chao Deng, Chen Zhao, Jingjing Yang, Junlan Feng, Kexin Yang, Xiaofan Liu, Xing Wang, Zhendong Wang, Zhiyan Song","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T09:02:04Z","title":"TReB: A Comprehensive Benchmark for Evaluating Table Reasoning Capabilities of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18421","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":"sha256:9e6d91c623af499df621d6ff8aa3787c774c87961b87a706637847d6becd70e5","target":"record","created_at":"2026-07-05T11:36:33Z","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":"f3ad9c05b1378f9c0a9c12e54996e14b551711e8b05d578e8a973477e74d519a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T09:02:04Z","title_canon_sha256":"25258c069a575d203368a6a58c06b00c2a7951010f3e17ca694a837765b7d57a"},"schema_version":"1.0","source":{"id":"2506.18421","kind":"arxiv","version":2}},"canonical_sha256":"4071d954a3df527a4946241979cb72b2b070dbd7bb4e89f0b10f2bba809c1207","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4071d954a3df527a4946241979cb72b2b070dbd7bb4e89f0b10f2bba809c1207","first_computed_at":"2026-07-05T11:36:33.456874Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:36:33.456874Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6iROA2RLBr1ZW7QTe/QE2+uyGOId2LftJLUXlHcjBWDYXODrKN88BrfZDg683PSRazeMnOA9HmF+mIqtAQQcCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:36:33.457393Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.18421","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e6d91c623af499df621d6ff8aa3787c774c87961b87a706637847d6becd70e5","sha256:ad6e9217b048e9e71de76e958a95ac49a96792acb80f61e8bcf9c44d5ccc3700"],"state_sha256":"e20636a47501761da26620cbc796fc3b1021355820ceaa4275571faafb3459db"}