When Chain-of-Thought Backfires: Evaluating Prompt Sensitivity in Medical Language Models

This work evaluates how chain-of-thought prompting strategies can paradoxically degrade the performance of medical language models, revealing unexpected prompt sensitivity in clinical reasoning tasks.

Recommended citation: Sadanandan, B. (2026). When Chain-of-Thought Backfires: Evaluating Prompt Sensitivity in Medical Language Models.