Title: ClearFairy: Capturing Creative Workflows through Decision Structuring, In-Situ Questioning, and Rationale Inference Show Chinese title

Title: ClearFairy：通过决策结构、现场提问和推理推断捕捉创意流程

Abstract: Capturing professionals' decision-making in creative workflows is essential for reflection, collaboration, and knowledge sharing, yet existing methods often leave rationales incomplete and implicit decisions hidden. To address this, we present CLEAR framework that structures reasoning into cognitive decision steps-linked units of actions, artifacts, and self-explanations that make decisions traceable. Building on this framework, we introduce ClearFairy, a think-aloud AI assistant for UI design that detects weak explanations, asks lightweight clarifying questions, and infers missing rationales to ease the knowledge-sharing burden. In a study with twelve creative professionals, 85% of ClearFairy's inferred rationales were accepted, increasing strong explanations from 14% to over 83% of decision steps without adding cognitive demand. The captured steps also enhanced generative AI agents in Figma, yielding next-action predictions better aligned with professionals and producing more coherent design outcomes. For future research on human knowledge-grounded creative AI agents, we release a dataset of captured 417 decision steps. Show Chinese abstract

Abstract: 捕捉专业人员在创意工作流中的决策过程对于反思、协作和知识共享至关重要，但现有方法常常导致理由不完整，隐性决策被隐藏。 为了解决这个问题，我们提出了CLEAR框架，该框架将推理结构化为与行动、成果和自我解释相关联的认知决策步骤，使决策可追溯。 在此框架的基础上，我们介绍了ClearFairy，这是一种用于UI设计的思考 aloud AI助手，它可以检测薄弱的解释，提出轻量级澄清问题，并推断缺失的理由，以减轻知识共享的负担。 在一项针对十二名创意专业人士的研究中，85%的ClearFairy推断的理由被接受，使强解释从14%增加到超过83%的决策步骤，而没有增加认知负担。 捕获的步骤还增强了Figma中的生成式AI代理，产生了更符合专业人员的下一步预测，并产生了更连贯的设计结果。 对于未来关于人类知识基础的创意AI代理的研究，我们发布了包含417个决策步骤的数据集。