Papers › OpenMTB-Audit: Exposing Over-Refusal and Clinical Expert Perspectives in LLM-Based...
OpenMTB-Audit: Exposing Over-Refusal and Clinical Expert Perspectives in LLM-Based Molecular Tumor Board Safety Evaluation
Negin Ashrafi, Jia Luo, Stacey M. Frumm, Roxana Daneshjou
Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).
Molecular tumor boards integrate genomic findings, clinical context, and therapeutic evidence to support precision oncology. As AI enters this workflow, a key safety challenge is distinguishing truly unsupported recommendations from evidence-supported options that still require oncologist review because of incomplete information, poor ECOG performance status, or other clinical caveats. We introduce OpenMTB-Audit, an open-source benchmark of 500 synthetic non-small cell lung cancer cases spanning five adversarial error categories and four safety labels: Supported, Partially Supported, Unsupported, and Insufficient Information. Across eight large language model configurations, we identify pervasive over-refusal: all LLM configurations failed to retain the Partially Supported label in 83.3-100% of true Partially Supported cases, achieving high aggregate safety scores through label collapse rather than clinically calibrated reasoning. To address this limitation, we developed MTB-AuditAgent, a deterministic seven-module framework separating evidence verification, missing-information detection, safety classification, and abstention. It reduces over-refusal to 6.7% and achieves 91.2% accuracy (95% CI: 88.6-93.6%). A two-oncologist annotation study found disagreement concentrated at the boundary between information sufficiency and treatment optimization, underscoring the need to preserve clinically meaningful distinctions.
In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.
For agents, on Syntology's MCP service (how to connect):
get_citation_path(paper_1="2610.01497", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2610.01497")
Code
Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Syntology holds the repository link but has not harvested or run code from it.
Results from the paper
The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2610.01497, was issued in October 2026, after that date, so the archive has no leaderboard rows for it.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections