RAmBLA: A Framework for Evaluating the Reliability of LLMs as Assistants in the Biomedical Domain
This work addresses the reliability of LLMs for high-impact biomedical applications, but it is incremental as it focuses on evaluation rather than new methods.
The authors tackled the problem of assessing the reliability of large language models (LLMs) as assistants in biomedicine by introducing the RAmBLA framework, evaluating four state-of-the-art models on tasks like prompt robustness and hallucination reduction, with results measured via semantic similarity to ground truth.
Large Language Models (LLMs) increasingly support applications in a wide range of domains, some with potential high societal impact such as biomedicine, yet their reliability in realistic use cases is under-researched. In this work we introduce the Reliability AssesMent for Biomedical LLM Assistants (RAmBLA) framework and evaluate whether four state-of-the-art foundation LLMs can serve as reliable assistants in the biomedical domain. We identify prompt robustness, high recall, and a lack of hallucinations as necessary criteria for this use case. We design shortform tasks and tasks requiring LLM freeform responses mimicking real-world user interactions. We evaluate LLM performance using semantic similarity with a ground truth response, through an evaluator LLM.