3 papers
cs.CL2026
Perturbation-Based Data Selection for Medical Question Answering
Yuexin Wu, Dayou Yu, Vasile Rus
Selecting a compact training subset is an important step in adapting medical language models from existing question-answering banks. We investigate perturbation-based data selectio…
cs.CL2026
An Empirical Study of Development-Tuned Representation Sensitivity for Medical QA Data Selection
Yuexin Wu, Dayou Yu, Vasile Rus
We study medical question-answering (QA) data selection when a developer already has questions, labels, and auxiliary explanatory texts but can retain only examples for a final…
cs.AI2026
Safe to Stop? Risk-Constrained Stopping for Sequential Clinical Diagnosis Agents
Yuexin Wu, Vasile Rus
Clinical diagnosis agents must decide not only what test to request next, but also when to diagnose or defer. Existing agent benchmarks largely evaluate accuracy after fixed or unc…