MedCoG：通过元认知调节最大化医学推理中的LLM推理密度
MedCoG: Maximizing LLM Inference Density in Medical Reasoning via Meta-Cognitive Regulation
AI总结：
提出MedCoG框架，利用元认知评估动态调节知识使用，以缓解推理扩展定律下的收益递减，提升推理效率与准确性。
摘要：
大型语言模型（LLM）在复杂医学推理中展现出强大潜力，但在推理扩展定律下面临收益递减。现有研究通过添加各种知识类型增强LLM，但额外成本转化为准确性的效果尚不明确。本文探索LLM的元认知（即对其自身认知状态的自我评估）如何调节推理过程。具体而言，我们提出MedCoG，一种带有知识图谱的医学元认知智能体，其中对任务复杂性、熟悉度和知识密度的元认知评估动态调节程序性、情景性和事实性知识的利用。这种以LLM为中心的按需推理旨在通过（1）避免无差别扩展以降低成本，（2）过滤干扰知识以提高准确性，来缓解扩展定律下的收益递减。为验证这一点，我们经验性地刻画了扩展曲线，并引入推理密度来量化推理效率。实验表明MedCoG在五个医学基准困难集上的有效性和高效性，实现了6.2倍的推理密度。此外，Oracle研究凸显了元认知调节的巨大潜力。
英文摘要：
Large Language Models (LLMs) have shown strong potential in complex medical reasoning yet face diminishing gains under inference scaling laws. While existing studies augment LLMs with various knowledge types, it remains unclear how effectively the additional costs translate into accuracy. In this paper, we explore how meta-cognition of LLMs, i.e., their self-assessment of their own cognitive states, can regulate the reasoning process. Specifically, we propose MedCoG, a Medical Meta-Cognition Agent with Knowledge Graph, where the meta-cognitive assessments of task complexity, familiarity, and knowledge density dynamically regulate utilization of procedural, episodic, and factual knowledge. The LLM-centric on-demand reasoning aims to mitigate the diminishing returns under scaling law by (1) reducing costs via avoiding indiscriminate scaling, (2) improving accuracy via filtering out distractive knowledge. To validate this, we empirically characterize the scaling curve and introduce inference density to quantify inference efficiency. Experiments demonstrate the effectiveness and efficiency of MedCoG on five hard sets of medical benchmarks, yielding 6.2x inference density. Furthermore, the Oracle study highlights the significant potential of meta-cognitive regulation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
