Large language models (LLMs) present a valuable technology for various
applications in healthcare, but their tendency to hallucinate introduces
unacceptable uncertainty in critical decision-making situations. Human-AI
collaboration (HAIC) can mitigate this uncertainty by combining human and AI
strengths for better outcomes. This paper presents a novel guided deferral
system that provides intelligent guidance when AI defers cases to human
decision-makers. We leverage LLMs' verbalisation capabilities and internal
states to create this system, demonstrating that fine-tuning smaller LLMs with
data from larger models enhances performance while maintaining computational
efficiency. A pilot study showcases the effectiveness of our deferral system.

通过利用大型语言模型（LLMs）的口述能力和内部状态，该研究提出了一种新的引导性延期系统，结合人工智能和人类决策者的优势，以减少 LLMs 在关键决策时产生的不确定性，并证明通过利用大型模型的数据对较小的 LLMs 进行微调可以提高性能并保持计算效率。一项初步研究展示了该延期系统的有效性。