AUG 8, 2026 · JOURNAL ARTICLE
Beyond Accuracy: A Mixed-Methods Audit of Chain-of-Thought Failures in LLM-Based COVID-19 Vaccine Stance Detection
Journal of Medical Systems
This is a single mixed-methods technical audit of LLM performance on a specific task (stance classification) with no clinical trial data, no patient outcomes, and no direct evidence of impact on clinical practice or public health decision-making.
Reported
Accuracy (o4-mini)0.819
Accuracy (Gemini)0.799
Macro-F1 (o4-mini)0.8