LLM Sensitivity Evaluation Framework for Clinical Diagnosis
This work addresses the need for more reliable and sensitive LLMs in clinical diagnosis, which is crucial for enhancing trust and practical application in healthcare, though it is incremental as it builds on existing evaluation frameworks.
The paper tackled the problem of evaluating large language models' sensitivity to key medical information in clinical diagnosis, finding that current models like GPT-4 and Gemini show limitations in maintaining sensitivity to subtle variations that affect diagnostic decisions.
Large language models (LLMs) have demonstrated impressive performance across various domains. However, for clinical diagnosis, higher expectations are required for LLM's reliability and sensitivity: thinking like physicians and remaining sensitive to key medical information that affects diagnostic reasoning, as subtle variations can lead to different diagnosis results. Yet, existing works focus mainly on investigating the sensitivity of LLMs to irrelevant context and overlook the importance of key information. In this paper, we investigate the sensitivity of LLMs, i.e. GPT-3.5, GPT-4, Gemini, Claude3 and LLaMA2-7b, to key medical information by introducing different perturbation strategies. The evaluation results highlight the limitations of current LLMs in remaining sensitive to key medical information for diagnostic decision-making. The evolution of LLMs must focus on improving their reliability, enhancing their ability to be sensitive to key information, and effectively utilizing this information. These improvements will enhance human trust in LLMs and facilitate their practical application in real-world scenarios. Our code and dataset are available at https://github.com/chenwei23333/DiagnosisQA.