{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.24145", "kind": "paper", "url": "https://arxiv.org/abs/2606.24145"}], "authors": ["Saba A. Farahani", "Hung Cao", "Ramesh Jain", "Amir M. Rahmani"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.24145", "source_url": "https://arxiv.org/abs/2606.24145"}, "categories": ["Biology & Drug Discovery", "Factuality", "Language & Knowledge", "cs.AI"], "description": {"en": "T2D-Bench evaluates LLM outputs for type 2 diabetes against evidence constraints using a multi-layer clinical-lifestyle knowledge graph, covering diagnosis, medication safety, and lifestyle conflicts across 100 structured vignettes."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.24145", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.24145", "title": "T2D-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of California, Irvine"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.24145", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.24145", "source_url": "https://openalex.org/W7165728378"}, "key": "claire-radar:2606.24145", "modality": null, "name": "T2D-Bench", "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_t2d-bench_6b92ee7c", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "736fbc75e43f332c2b9c5673710fcd10b8437c3c5cfdffd69aca2e339053c8aa", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.24145"}, "publisher": null, "released": "2026-06-23", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.24145", "source_url": "https://arxiv.org/abs/2606.24145"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-24145", "source": "claire_radar", "source_benchmark_id": "2606.24145", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Biology & Drug Discovery"], "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/2606.24145", "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": ["Factuality"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "15c2d1bbec2fdc0844a712ad7ee5c89bf538628056cfea5a1ff4d76fdc364dbd", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "T2D-Bench evaluates LLM outputs for type 2 diabetes against evidence constraints using a multi-layer clinical-lifestyle knowledge graph, covering diagnosis, medication safety, and lifestyle conflicts across 100 structured vignettes.", "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 present T2D-Bench, a reproducible benchmark and evidence-gated evaluation framework for testing whether LLM outputs satisfy explicit, graph-checkable evidence requirements."}, "familyId": "bmf_ece4accf1aab", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_t2d-bench_6b92ee7c", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.24145", "paper": "https://arxiv.org/abs/2606.24145", "pdf": "https://arxiv.org/pdf/2606.24145", "project": null, "report": "https://arxiv.org/abs/2606.24145"}, "metrics": [], "motivation": "Large language models (LLMs) can produce clinically fluent recommendations for type 2 diabetes while failing to satisfy guideline constraints or explicitly justify lifestyle-related glycemic claims.", "name": "T2D-Bench", "oneLine": "T2D-Bench evaluates LLM outputs for type 2 diabetes against evidence constraints using a multi-layer clinical-lifestyle knowledge graph, covering diagnosis, medication safety, and lifestyle conflicts across 100 structured vignettes.", "paperTitle": "T2D-Bench: Evidence-Gated Evaluation of LLM Outputs for Type 2 Diabetes Using a Multi-Layer Clinical-Lifestyle Knowledge Graph", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": "7 pages, 2 figures, 2 tables. Accepted as a poster at AMIA 2026 Annual Symposium", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.24145", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "as a poster at AMIA 2026 Annual Symposium", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-23", "source": {"authors": ["Saba A. Farahani", "Hung Cao", "Ramesh Jain", "Amir M. Rahmani"], "categories": ["cs.AI"], "id": "2606.24145", "title": "T2D-Bench: Evidence-Gated Evaluation of LLM Outputs for Type 2 Diabetes Using a Multi-Layer Clinical-Lifestyle Knowledge Graph", "type": "arxiv", "url": "https://arxiv.org/abs/2606.24145"}, "sourceUpdatedAt": "2026-06-23T05:02:21Z", "topics": ["cs.AI"], "venueAttempts": [{"decisionRaw": "7 pages, 2 figures, 2 tables. Accepted as a poster at AMIA 2026 Annual Symposium", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "7 pages, 2 figures, 2 tables. Accepted as a poster at AMIA 2026 Annual Symposium", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.24145"}], "reviewStatus": "accepted", "venueName": "as a poster at AMIA 2026 Annual Symposium"}], "whyItMatters": "Addresses the gap in evaluating whether LLM recommendations satisfy explicit clinical guidelines and justify lifestyle-related glycemic claims, providing a mechanism to detect unsupported omissions and improve verifier-level compliance."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
