{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OQRT2IZ3JE6B2DJYQ7PXEFB4TV","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":"ad4fade8fa572c4da81337afcc7d04435a0a912f05a09d237893b88cb27683a3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-13T13:52:15Z","title_canon_sha256":"09742a9b82406b39d998d6fee2cefab277720d9a4a06955f958d0a3493fae2ed"},"schema_version":"1.0","source":{"id":"2310.09107","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.09107","created_at":"2026-07-05T10:51:18Z"},{"alias_kind":"arxiv_version","alias_value":"2310.09107v2","created_at":"2026-07-05T10:51:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.09107","created_at":"2026-07-05T10:51:18Z"},{"alias_kind":"pith_short_12","alias_value":"OQRT2IZ3JE6B","created_at":"2026-07-05T10:51:18Z"},{"alias_kind":"pith_short_16","alias_value":"OQRT2IZ3JE6B2DJY","created_at":"2026-07-05T10:51:18Z"},{"alias_kind":"pith_short_8","alias_value":"OQRT2IZ3","created_at":"2026-07-05T10:51:18Z"}],"graph_snapshots":[{"event_id":"sha256:95c54fdfacb5e4f10fd579fccbd9d97e1768f457f6b0c9872142b3fd1ab2701b","target":"graph","created_at":"2026-07-05T10:51:18Z","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/2310.09107/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have shown significant general language understanding abilities. However, there has been a scarcity of attempts to assess the logical reasoning capacities of these LLMs, an essential facet of natural language understanding. To encourage further investigation in this area, we introduce GLoRE, a General Logical Reasoning Evaluation platform that not only consolidates diverse datasets but also standardizes them into a unified format suitable for evaluating large language models across zero-shot and few-shot scenarios. Our experimental results show that compared to the","authors_text":"Hanmeng Liu, Ruoxi Ning, Xiaozhang Liu, Xiulai Li, Yiran Ding, Yue Zhang, Zhiyang Teng","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-13T13:52:15Z","title":"GLoRE: Evaluating Logical Reasoning of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.09107","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":"sha256:6bebafcc380ac57d3a385f812ecfbdb434628e7927db2c5926817c939db28519","target":"record","created_at":"2026-07-05T10:51:18Z","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":"ad4fade8fa572c4da81337afcc7d04435a0a912f05a09d237893b88cb27683a3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-13T13:52:15Z","title_canon_sha256":"09742a9b82406b39d998d6fee2cefab277720d9a4a06955f958d0a3493fae2ed"},"schema_version":"1.0","source":{"id":"2310.09107","kind":"arxiv","version":2}},"canonical_sha256":"74233d233b493c1d0d3887df72143c9d53914b98f1c449714f7f9af13a30aecc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74233d233b493c1d0d3887df72143c9d53914b98f1c449714f7f9af13a30aecc","first_computed_at":"2026-07-05T10:51:18.940143Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:51:18.940143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZmiOwdpJ4nHdj1TA0HrzSpzwqK6wTSqFESwhYFUjme1YGt6uJIhIsrXDcc9kUUnt/11i9M8IKz/6zcJ/LK4iBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:51:18.940642Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.09107","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6bebafcc380ac57d3a385f812ecfbdb434628e7927db2c5926817c939db28519","sha256:95c54fdfacb5e4f10fd579fccbd9d97e1768f457f6b0c9872142b3fd1ab2701b"],"state_sha256":"88e29e4a70fbac60304062b813d3de6d452bf010bd2455beb8dece223506e70e"}