DeCode: Decoupling Content and Delivery for Medical QA
It addresses the need for personalized medical responses for patients, though it is incremental as it adapts existing LLMs without new training.
The paper tackles the problem of large language models (LLMs) generating clinically correct but contextually misaligned answers in medical QA by introducing DeCode, a training-free framework that improves state-of-the-art performance from 28.4% to 49.8% on a benchmark.
Large language models (LLMs) exhibit strong medical knowledge and can generate factually accurate responses. However, existing models often fail to account for individual patient contexts, producing answers that are clinically correct yet poorly aligned with patients' needs. In this work, we introduce DeCode, a training-free, model-agnostic framework that adapts existing LLMs to produce contextualized answers in clinical settings. We evaluate DeCode on OpenAI HealthBench, a comprehensive and challenging benchmark designed to assess clinical relevance and validity of LLM responses. DeCode improves the previous state of the art from $28.4\%$ to $49.8\%$, corresponding to a $75\%$ relative improvement. Experimental results suggest the effectiveness of DeCode in improving clinical question answering of LLMs.