{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UMCJKFGFX6CMTICOFJV4CVI7UZ","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":"5eb11e07f25bd0748db537c933a93ff58b2e79a32e56ce1ef1e168a30bf47d89","cross_cats_sorted":["cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-15T06:42:00Z","title_canon_sha256":"c06d1df8e075f23e0b8498cd56b62f2269c0c39ab48e25d17418cf72f8bb374d"},"schema_version":"1.0","source":{"id":"2508.11252","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11252","created_at":"2026-07-05T11:54:21Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11252v1","created_at":"2026-07-05T11:54:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11252","created_at":"2026-07-05T11:54:21Z"},{"alias_kind":"pith_short_12","alias_value":"UMCJKFGFX6CM","created_at":"2026-07-05T11:54:21Z"},{"alias_kind":"pith_short_16","alias_value":"UMCJKFGFX6CMTICO","created_at":"2026-07-05T11:54:21Z"},{"alias_kind":"pith_short_8","alias_value":"UMCJKFGF","created_at":"2026-07-05T11:54:21Z"}],"graph_snapshots":[{"event_id":"sha256:864c11481442c4db9d20e6a8b0f9b71bd60f3666a8e845a5ebc10c4749d7064e","target":"graph","created_at":"2026-07-05T11:54:21Z","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/2508.11252/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Reasoning Models (LRMs) have demonstrated remarkable problem-solving abilities in mathematics, as evaluated by existing benchmarks exclusively on well-defined problems. However, such evaluation setup constitutes a critical gap, since a genuine intelligent agent should not only solve problems (as a math quiz solver), but also be able~to ask for information when the problems lack sufficient information, enabling proactivity in responding users' requests. To bridge such gap, we proposes a new dataset consisting of two types of incomplete problems with diverse contexts. Based on the dataset,","authors_text":"Bowen Qin, Chen Huang, Duanyu Feng, Wenqiang Lei, Xi Yang, Youcheng Huang","cross_cats":["cs.CL","cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-15T06:42:00Z","title":"Beyond Solving Math Quiz: Evaluating the Ability of Large Reasoning Models to Ask for Information"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11252","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":"sha256:5b46f85b13892368cbaaadb3a2dce1bf249e13d153d19d72df93bc80a46484ac","target":"record","created_at":"2026-07-05T11:54:21Z","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":"5eb11e07f25bd0748db537c933a93ff58b2e79a32e56ce1ef1e168a30bf47d89","cross_cats_sorted":["cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-15T06:42:00Z","title_canon_sha256":"c06d1df8e075f23e0b8498cd56b62f2269c0c39ab48e25d17418cf72f8bb374d"},"schema_version":"1.0","source":{"id":"2508.11252","kind":"arxiv","version":1}},"canonical_sha256":"a3049514c5bf84c9a04e2a6bc1551fa66c01f6ae10c036f541ad4798e27c6d7c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a3049514c5bf84c9a04e2a6bc1551fa66c01f6ae10c036f541ad4798e27c6d7c","first_computed_at":"2026-07-05T11:54:21.402483Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:54:21.402483Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fJnFgTgxUn83uQwqH1NUEcEoFBr7zWV7XhccoahwA7vQjMT58T1u9yUdMqZtjLc2ic+IDWX/K7uwa/fY4YjVCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:54:21.402900Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.11252","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b46f85b13892368cbaaadb3a2dce1bf249e13d153d19d72df93bc80a46484ac","sha256:864c11481442c4db9d20e6a8b0f9b71bd60f3666a8e845a5ebc10c4749d7064e"],"state_sha256":"374b86966686e0b399090862cd05819bca3599807381753113b46ae98c86591a"}