{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FJMVCPCPC53U3MFX67UGVG7PZK","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":"44314cf278a9c9ba4801a21ea3094a59ad9bfe22f75a1ef36290ee9de8404581","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-10T15:31:54Z","title_canon_sha256":"11cb8bbc3ac8f2e9746e5f2890cefdaf226b8fc121c60125c98d9ed34f630e82"},"schema_version":"1.0","source":{"id":"2502.06563","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06563","created_at":"2026-07-05T10:22:13Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06563v2","created_at":"2026-07-05T10:22:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06563","created_at":"2026-07-05T10:22:13Z"},{"alias_kind":"pith_short_12","alias_value":"FJMVCPCPC53U","created_at":"2026-07-05T10:22:13Z"},{"alias_kind":"pith_short_16","alias_value":"FJMVCPCPC53U3MFX","created_at":"2026-07-05T10:22:13Z"},{"alias_kind":"pith_short_8","alias_value":"FJMVCPCP","created_at":"2026-07-05T10:22:13Z"}],"graph_snapshots":[{"event_id":"sha256:448e48d007c285c77774919545215f117bea6e32f0fc4da354a99bc3c658501f","target":"graph","created_at":"2026-07-05T10:22:13Z","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/2502.06563/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"First-order logic (FOL) reasoning, which involves sequential deduction, is pivotal for intelligent systems and serves as a valuable task for evaluating reasoning capabilities, particularly in chain-of-thought (CoT) contexts. Existing benchmarks often rely on extensive human annotation or handcrafted templates, making it difficult to achieve the necessary complexity, scalability, and diversity for robust evaluation. To address these limitations, we propose a novel framework called ProverGen that synergizes the generative strengths of Large Language Models (LLMs) with the rigor and precision of ","authors_text":"Binyuan Hui, Bowen Li, Chengwen Qi, Conghui He, He Du, Jinwang Wu, Ren Ma, Yuanjun Laili","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-10T15:31:54Z","title":"Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06563","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":"sha256:79bf21dad775815441c374fe84e1c273b25f6886c9888b152161f28cc4ffd8a6","target":"record","created_at":"2026-07-05T10:22:13Z","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":"44314cf278a9c9ba4801a21ea3094a59ad9bfe22f75a1ef36290ee9de8404581","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-10T15:31:54Z","title_canon_sha256":"11cb8bbc3ac8f2e9746e5f2890cefdaf226b8fc121c60125c98d9ed34f630e82"},"schema_version":"1.0","source":{"id":"2502.06563","kind":"arxiv","version":2}},"canonical_sha256":"2a59513c4f17774db0b7f7e86a9befcab6542a41b7c017a38226699145cf3fbd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2a59513c4f17774db0b7f7e86a9befcab6542a41b7c017a38226699145cf3fbd","first_computed_at":"2026-07-05T10:22:13.036387Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:22:13.036387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vTG0rWrS4Oyy48W+9RlA/V8sJEUIbWpjKb+hWALW3uuK5T6/M6LWRhM6j7iDGPXoCmqR7L4Mhnki8rDltDu9Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:22:13.036908Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06563","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79bf21dad775815441c374fe84e1c273b25f6886c9888b152161f28cc4ffd8a6","sha256:448e48d007c285c77774919545215f117bea6e32f0fc4da354a99bc3c658501f"],"state_sha256":"be60682367e27af414366c645501183a8dc67c776fe4033f7667cc2173624fbc"}