EHR-derived standardized patients and dual-track evaluation reveal LLMs trail clinicians by 37.28 points on full psychiatric encounters, with mental-status assessment the main bottleneck.
arXiv preprint arXiv:2601.03023 , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
A parallel multi-turn medical dialogue dataset spanning English and nine Indic languages is created from synthetic consultations to enable personalized AI healthcare interactions.
MedConceal provides 300 cases and a simulator that withholds hidden concerns to evaluate confirmation and intervention in medical dialogue, finding frontier models vary on surfacing concerns while humans outperform on guiding patients to care plans.
MedDialBench shows LLMs suffer 1.7-3.4x larger diagnostic accuracy drops from patients fabricating symptoms than withholding them, with fabrication driving super-additive interaction effects across models.
citing papers explorer
-
MentalHospital: A Virtual Environment for Evaluating Psychiatric Clinical Encounters
EHR-derived standardized patients and dual-track evaluation reveal LLMs trail clinicians by 37.28 points on full psychiatric encounters, with mental-status assessment the main bottleneck.
-
IndicMedDialog: A Parallel Multi-Turn Medical Dialogue Dataset for Accessible Healthcare in Indic Languages
A parallel multi-turn medical dialogue dataset spanning English and nine Indic languages is created from synthetic consultations to enable personalized AI healthcare interactions.
-
MedConceal: A Benchmark for Clinical Hidden-Concern Reasoning Under Partial Observability
MedConceal provides 300 cases and a simulator that withholds hidden concerns to evaluate confirmation and intervention in medical dialogue, finding frontier models vary on surfacing concerns while humans outperform on guiding patients to care plans.
-
MedDialBench: Benchmarking LLM Diagnostic Robustness under Parametric Adversarial Patient Behaviors
MedDialBench shows LLMs suffer 1.7-3.4x larger diagnostic accuracy drops from patients fabricating symptoms than withholding them, with fabrication driving super-additive interaction effects across models.