摘要
大语言模型虽展现出初步推理能力，但仍局限于语言模拟，缺乏时间连续性与现实锚定。本文提出一种互补方法，将推理视为分布于人与模型间的关系过程。基于“系统 2
AI 推荐理由
论文核心提出人机协作的反思性推理框架，将推理视为交互过程而非模型内部能力。
研究机构
Faculty of Medicine, Lund University, Sweden
Department of Economics, Lund University School of Economics and Management, Sweden
Department of Health Services Research and Policy, London School of Hygiene & Tropical Medicine, UK
论文信息