arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.15331cs.HC
收藏

DesignerlyLoop: 通过精选推理形成设计意图以实现人-大语言模型对齐

DesignerlyLoop: Forming Design Intent through Curated Reasoning for Human-LLM Alignment

Hong Kong University of Science and Technology(香港科技大学);Central South University(中南大学);Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州))

Anqi Wang, Zhengyi Li, Xin Tong, Pan Hui

AI总结:

DesignerlyLoop通过精选推理提升人-大语言模型在设计任务中的对齐,改进设计质量和创造力。

摘要:

近期大型语言模型(LLMs)在设计任务中展现出潜力,但仍然存在根本性的不匹配:设计思维需要迭代的意图形成,而LLMs将输入视为完整的规范。这挑战了意图形成,其中设计师必须通过探索逐步完善理解。现有工具要么牺牲探索的灵活性以换取结构的稳定性,要么让推理隐含,无法支持人-LLM对齐。通过与八位设计师的形成研究,我们引入了精选推理,使设计师能够显式地检查、重组和选择性地再生LLM的推理结构。我们提出了DesignerlyLoop,通过两层结构将设计意图与LLM推理分离。与20位设计师的实验表明,精选推理显著提高了设计质量和创造力。我们的工作贡献了一种新的交互范式,将LLMs从内容生成者转变为创意设计中的结构推理伙伴。

英文摘要:

Recent large language models (LLMs) show promise in design tasks, yet a fundamental misalignment persists: design thinking requires iterative intent formulation, while LLMs treat inputs as complete specifications. This challenges design intent formulation, where designers must progressively refine understanding through exploration. Existing tools either sacrifice exploratory flexibility for structural stability or leave reasoning implicit, failing to support human-LLM alignment. Through a formative study with eight designers, we introduce curated reasoning-enabling designers to explicitly inspect, reorganize, and selectively regenerate LLM reasoning structures. We present DesignerlyLoop, implementing this through a two-layer structure separating design intent from LLM reasoning. A study with 20 designers demonstrates that curated reasoning significantly improves design quality and creativity. Our work contributes a novel interaction paradigm for human-LLM alignment, transforming LLMs from content generators into structured reasoning partners in creative design.

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。

↑