{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VJJEAVS3O43M6MYZWXSLQJUR4Q","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ce13c24ae308214855529f935ad014689bb38980293b1849ebe4310aec6b9f92","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-13T17:38:22Z","title_canon_sha256":"d9b9d1f4ff20a5e1d53a364325e4af3cdb25b4005159cc289fea9c77bd72e062"},"schema_version":"1.0","source":{"id":"2607.11849","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.11849","created_at":"2026-07-14T02:22:35Z"},{"alias_kind":"arxiv_version","alias_value":"2607.11849v1","created_at":"2026-07-14T02:22:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.11849","created_at":"2026-07-14T02:22:35Z"},{"alias_kind":"pith_short_12","alias_value":"VJJEAVS3O43M","created_at":"2026-07-14T02:22:35Z"},{"alias_kind":"pith_short_16","alias_value":"VJJEAVS3O43M6MYZ","created_at":"2026-07-14T02:22:35Z"},{"alias_kind":"pith_short_8","alias_value":"VJJEAVS3","created_at":"2026-07-14T02:22:35Z"}],"graph_snapshots":[{"event_id":"sha256:007c6ec89b13ef776c6b42f26cad1a7270528dcd9782edc8bc1d24226de94299","target":"graph","created_at":"2026-07-14T02:22:35Z","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/2607.11849/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have achieved remarkable performance on high-school and olympiad-style mathematics, yet their capabilities on advanced mathematics remain poorly understood. Existing benchmarks, however, fall short in both scope and evaluation granularity: they provide limited disciplinary coverage and often rely on final-answer correctness or coarse judgments, leaving the validity of the reasoning process inadequately assessed. To bridge this gap, we introduce AdvancedMathBench, a benchmark suite designed to evaluate advanced mathematical reasoning capabilities. Its core proof-gen","authors_text":"Disheng Xu, Haiteng Zhao, Kai Chen, Lingkai Kong, Shuang Sun, Shuya Zhao, Wenwei Zhang, Wenyong Huang, Xiaotian Zhang, Yan Wang, Yuzhe Gu, Zhicheng Xiong, Zijian Wu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-13T17:38:22Z","title":"AdvancedMathBench: A Benchmark Suite for Advanced Mathematical Proof Generation and Verification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.11849","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:ad3f977ac528d15047f60b2f0a8e82d5e26fab09cfcf70560d6527fc0be282d5","target":"record","created_at":"2026-07-14T02:22:35Z","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":"ce13c24ae308214855529f935ad014689bb38980293b1849ebe4310aec6b9f92","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-13T17:38:22Z","title_canon_sha256":"d9b9d1f4ff20a5e1d53a364325e4af3cdb25b4005159cc289fea9c77bd72e062"},"schema_version":"1.0","source":{"id":"2607.11849","kind":"arxiv","version":1}},"canonical_sha256":"aa5240565b7736cf3319b5e4b82691e40971e8f3eaed27a3079a53d5cc77e65c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa5240565b7736cf3319b5e4b82691e40971e8f3eaed27a3079a53d5cc77e65c","first_computed_at":"2026-07-14T02:22:35.960607Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-14T02:22:35.960607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ajZTbvi/R53McJaq6XD6SjuU/Wourfo+zAxsrCzhlm6qQccpd/UiISzsKRZQcNLl/GN+BQ26iZieiU1eqPOmDw==","signature_status":"signed_v1","signed_at":"2026-07-14T02:22:35.961495Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.11849","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ad3f977ac528d15047f60b2f0a8e82d5e26fab09cfcf70560d6527fc0be282d5","sha256:007c6ec89b13ef776c6b42f26cad1a7270528dcd9782edc8bc1d24226de94299","sha256:d3c2efdcb0e99986321a1b681de2d46f9cb0797a1c1ccaf6980b45b12e922894"],"state_sha256":"2d81a56f215d0645afb140e145d35c8c41a1d830113ec2bb378e804cac0458d6"}