构建四维框架，揭示大模型在质性分析中的认知层级差异。
Bridging Human Interpretation and Machine Representation: A Landscape of Qualitative Data Analysis in the LLM Era
- 提出4×4认知层次矩阵，区分意义建构与建模深度。
- 发现现有系统多停留于描述性层面，缺乏理论推演与动态建模。
- 倡导透明化、可选的解释与建模机制，适配研究者需求。
大语言模型正越来越多地用于支持质性研究，但现有系统输出差异显著——从忠实于原始痕迹的摘要，到经理论中介的解释与系统模型不一而足。为明确这些差异，我们提出一个4×4的认知景观，横跨四个意义建构层次（描述性、分类性、诠释性、理论性）与四个建模层次（静态结构、阶段/时间线、因果路径、反馈动态）。将该框架应用于既有大模型自动化研究，发现其普遍偏向低阶意义建构与低承诺表征，极少可靠实现诠释性/理论性推理或动态建模。基于此差距，我们提出一项议程：推动大模型系统在解释与建模上的承诺显式化、可选择且可调控。
原文摘要 · Abstract (English)
LLMs are increasingly used to support qualitative research, yet existing systems produce outputs that vary widely--from trace-faithful summaries to theory-mediated explanations and system models. To make these differences explicit, we introduce a 4$\times$4 landscape crossing four levels of meaning-making (descriptive, categorical, interpretive, theoretical) with four levels of modeling (static structure, stages/timelines, causal pathways, feedback dynamics). Applying the landscape to prior LLM-based automation highlights a strong skew toward low-level meaning and low-commitment representations, with few reliable attempts at interpretive/theoretical inference or dynamical modeling. Based on the revealed gap, we outline an agenda for applying and building LLM-systems that make their interpretive and modeling commitments explicit, selectable, and governable.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。