{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.03858", "kind": "paper", "url": "https://arxiv.org/abs/2606.03858"}], "authors": ["Zetian Ouyang", "Linlin Wang", "Gerard de Melo", "Liang He"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.03858", "source_url": "https://arxiv.org/abs/2606.03858"}, "categories": ["Language & Knowledge", "Mathematics & Formal Science", "cs.AI"], "description": {"en": "PyraMathBench is a hierarchical benchmark with 32,505 questions derived from math word problems, evaluating numerical reasoning and mathematical capabilities across cognitive aspects and modalities."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.03858", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.03858", "title": "PyraMathBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["East China Normal University", "Hasso Plattner Institute", "University of Potsdam"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.03858", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.03858", "source_url": "https://openalex.org/W7163370577"}, "key": "claire-radar:2606.03858", "modality": null, "name": "PyraMathBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_pyramathbench_14f9eba5", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "ff14ee9cb33996e3f73026040fb85dc39792f65ec73a3633554af02c934a407c", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.03858"}, "publisher": null, "released": "2026-06-02", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.03858", "source_url": "https://arxiv.org/abs/2606.03858"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-03858", "source": "claire_radar", "source_benchmark_id": "2606.03858", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Mathematics & Formal Science"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.03858", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "de8d01ef9aab45b08c7c67582fd4ecabaa250fb9134f25d9d1a1907fd6bbd4d7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "PyraMathBench is a hierarchical benchmark with 32,505 questions derived from math word problems, evaluating numerical reasoning and mathematical capabilities across cognitive aspects and modalities.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce PyraMathBench, a comprehensive hierarchical benchmark with 32,505 questions derived from 7,404 math word problems, spanning 4 key cognitive aspects, 14 subcategories, and 2 modalities."}, "familyId": "bmf_560f968ae5dd", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_pyramathbench_14f9eba5", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.03858", "paper": "https://arxiv.org/abs/2606.03858", "pdf": "https://arxiv.org/pdf/2606.03858", "project": null, "report": "https://arxiv.org/abs/2606.03858"}, "metrics": [], "motivation": "Despite the pivotal role of numerical reasoning as the cornerstone of mathematical capabilities in large language models (LLMs) across applications, few benchmarks evaluate LLMs by integrating numerical processing and mathematical reasoning, hindering the interpretability of failures in math tasks.", "name": "PyraMathBench", "oneLine": "PyraMathBench is a hierarchical benchmark with 32,505 questions derived from math word problems, evaluating numerical reasoning and mathematical capabilities across cognitive aspects and modalities.", "paperTitle": "PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models", "primaryDomain": "Mathematics & Formal Science", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.03858", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-02", "source": {"authors": ["Zetian Ouyang", "Linlin Wang", "Gerard de Melo", "Liang He"], "categories": ["cs.AI"], "id": "2606.03858", "title": "PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2606.03858"}, "sourceUpdatedAt": "2026-06-02T16:32:53Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "No public artifacts or reuse path are provided, and the benchmark lacks a clear scoring contract for independent runs."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
