{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.11702", "kind": "paper", "url": "https://arxiv.org/abs/2606.11702"}, {"id": "gh:ivul-kaust/medcta", "kind": "repo", "url": "https://github.com/IVUL-KAUST/MedCTA"}, {"id": "hf:ivul-kaust/medcta", "kind": "dataset", "url": "https://huggingface.co/datasets/IVUL-KAUST/MedCTA"}, {"kind": "website", "url": "https://ivul-kaust.github.io/MedCTA/"}], "authors": ["Tajamul Ashraf", "Hyewon Jeong", "Fida Mohammad Thoker", "Bernard Ghanem"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.11702", "source_url": "https://arxiv.org/abs/2606.11702"}, "categories": ["Biology & Drug Discovery", "Multimodal"], "description": {"en": "MedCTA evaluates clinical tool agents on 107 clinician-verified, step-implicit tasks with multimodal inputs (radiology images, pathology slides, reports) and 5 deployed tools. Metrics cover tool selection, argument validity, execution stability, trajectory fidelity, and outcome quality."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.11702", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.11702", "title": "MedCTA"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["King Abdullah University of Science and Technology", "Massachusetts Institute of Technology"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.11702", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.11702", "source_url": "https://openalex.org/W7164360100"}, "key": "claire-radar:2606.11702", "modality": null, "name": "MedCTA", "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_medcta_84809bbe", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "a70512b853c23d9ca28613ef272e0d355cf20950cf9f0a3a95917549f39115ca", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.11702"}, "publisher": null, "released": "2026-06-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.11702", "source_url": "https://arxiv.org/abs/2606.11702"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-11702", "source": "claire_radar", "source_benchmark_id": "2606.11702", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Biology & Drug Discovery"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/IVUL-KAUST/MedCTA", "githubScope": "benchmark_repo", "githubStars": 8, "hfDailySubmittedAt": null, "hfDataset": "IVUL-KAUST/MedCTA", "hfDatasetDownloads": 305, "hfDatasetLikes": 2, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.11702", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-24", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-24", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "capabilities": [], "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": "1d3af7081fe9ba9810369fee5a9b4369e7e263d419ecf502a88211a080190291", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MedCTA evaluates clinical tool agents on 107 clinician-verified, step-implicit tasks with multimodal inputs (radiology images, pathology slides, reports) and 5 deployed tools. Metrics cover tool selection, argument validity, execution stability, trajectory fidelity, and outcome quality.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce MedCTA, a benchmark for evaluating medical tool agents on clinician-validated, step-implicit tasks grounded in realistic multimodal clinical inputs, including radiology images, pathology slides, and reports."}, "familyId": "bmf_f4b829da8324", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/IVUL-KAUST/MedCTA", "url": "https://github.com/IVUL-KAUST/MedCTA"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/IVUL-KAUST/MedCTA", "url": "https://huggingface.co/papers/2606.11702"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/IVUL-KAUST/MedCTA", "url": "https://ivul-kaust.github.io/MedCTA/"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/IVUL-KAUST/MedCTA", "url": "https://arxiv.org/abs/2606.11702"}], "scope": "benchmark_repo", "url": "https://github.com/IVUL-KAUST/MedCTA", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_medcta_84809bbe", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": "https://github.com/IVUL-KAUST/MedCTA", "data": "https://huggingface.co/datasets/IVUL-KAUST/MedCTA", "hfPaper": "https://huggingface.co/papers/2606.11702", "paper": "https://arxiv.org/abs/2606.11702", "pdf": "https://arxiv.org/pdf/2606.11702", "project": "https://ivul-kaust.github.io/MedCTA/", "report": "https://arxiv.org/abs/2606.11702"}, "metrics": [], "motivation": "To make clinically grounded decisions, medical AI agents are expected to go beyond simple recognition and be capable of tool retrieval, evidence acquisition, and integration.", "name": "MedCTA", "oneLine": "MedCTA evaluates clinical tool agents on 107 clinician-verified, step-implicit tasks with multimodal inputs (radiology images, pathology slides, reports) and 5 deployed tools. Metrics cover tool selection, argument validity, execution stability, trajectory fidelity, and outcome quality.", "paperTitle": "MedCTA: A Benchmark for Clinical Tool Agents", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.11702", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "IVUL-KAUST", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://ivul-kaust.github.io/MedCTA/"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-10", "source": {"authors": ["Tajamul Ashraf", "Hyewon Jeong", "Fida Mohammad Thoker", "Bernard Ghanem"], "categories": ["cs.CV", "cs.AI", "cs.CL"], "id": "2606.11702", "title": "MedCTA: A Benchmark for Clinical Tool Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2606.11702"}, "sourceUpdatedAt": "2026-06-10T06:26:52Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Fills a gap in medical AI evaluation by going beyond single-turn QA to test agentic tool use, planning, and reliability in real-world clinical workflows. Useful for auditing and advancing trustworthy medical AI agents."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
