cs.CL提交于 2026-10-27已译

是什么让问题充满好奇？衡量人机交互中的模型好奇心

What Makes a Question Curious? Measuring Model Curiosity in Human-LLM Interactions

Angana Borah · Rachana Ramchandar · Inderjeet Nair · Rada Mihalcea

中文摘要

大型语言模型日益成为日常交流中的伙伴。在这些交互中，提供有用帮助不仅需要回答问题，还需要提出恰当问题，以揭示知识缺口、隐含假设和替代方案。研究将此类认知性提问定义为模型好奇心(model curiosity)，并提出一套基于理论的分类体系，用于区分面向好奇的问题与模型的一般性提问。在四个对话数据集中，面向好奇的问题仍然较少，仅占全部问题的 17–28%，但这类问题与更具探索性、认知内容更丰富的后续用户回应相关，且不同数据集呈现不同模式。进一步开展的人体研究发现，模型好奇心能够促进探索，尤其是在建议、创意和科研场景中。这些结果有助于构建能够更有选择性地提问的 LLM，从而支持更深入的探索与人机协作。

关键要点

01提出基于理论的问题分类体系，区分好奇型提问与一般模型提问。
02在四个对话数据集中，好奇型问题仅占全部问题的 17–28%。
03好奇型提问与更具探索性和认知丰富度的后续用户回应相关。
04人体研究显示，模型好奇心在建议、创意和科研场景中尤其有助于探索。
05研究结果可指导 LLM 更有选择性地提问，以促进深度探索与人机协作。

解读

尚无解读。

原始英文摘要

arXiv:2610.33950v1 Announce Type: new Abstract: Large language models increasingly act as partners in everyday communication. In these interactions, useful assistance may require not only answering, but also asking the right questions that surface knowledge gaps, assumptions, and alternatives. In this study, we refer to such epistemic question-asking as model curiosity, and introduce a theory-grounded taxonomy that distinguishes curiosity-oriented questions from generic model questioning. Across four conversation datasets, we find that curiosity-oriented questions remain rare (17-28% of all questions), but are associated with more exploratory and epistemically rich subsequent user responses, with varying patterns across datasets. We further conduct a human study to evaluate the usefulness of model curiosity, finding that curiosity can encourage exploration, especially in advice, creative, and research contexts. These findings have implications for building LLMs that ask more selectively to support deeper exploration and human-AI collaboration.