MedMisBench: Measuring Epistemic Resilience of LLMs Under Misleading Medical Context
Hongjian Zhou, Xinyu Zou, Jinge Wu, Sean Wu, Junchi Yu, Bradley Segal, Tobias Erich Niebuhr, Sara Amro, Michael Petrus, Sheikh Momin, Alexandra Cardoso Pinto, Rachel Niesen, Laura Sophie Wegner, Dhruv Darji, Jung Moses Koo, , David A. Clifton
The paper introduces MedMisBench, a benchmark to measure LLMs' ability to maintain correct medical judgment under misleading context, showing significant drops in accuracy when exposed to adversarial inputs.
bioRxiv · Jun 10, 2026 · PreprintBioengineering