The paper "Jemplate: Exploring LLM-Enhanced Templates for Idea Reflection" by Xiaotong (Tone) Xu explores the integration of large language models (LLMs) into design templates to enhance idea reflection and critical thinking. The authors highlight the challenges of using chat-based LLMs, such as the need for users to craft effective prompts and the tendency for LLM responses to be lengthy and error-prone. To address these issues, they propose an alternative approach using structured templates, which are already effective tools for helping novices organize and reflect on information. The study includes formative research with 52 students and five instructors, revealing that novice designers often lack guidance on how to use LLMs effectively. Based on these insights, the authors developed Jemplate, a digital whiteboard plugin that integrates LLM capabilities into design templates. Jemplate streamlines the collection and organization of user-generated content and LLM responses within the template structure, providing reflective questions and in-situ guidance to help users think critically and improve their ideas. A preliminary study with 8 novice designers showed that Jemplate's reflective questions and in-situ guidance improved their ability to think critically and enhance their ideas. The paper discusses the potential of LLM-enhanced templates to instigate critical reflection and provides design guidelines for integrating LLMs into templates, including leveraging template integration, precision prompting, and reflection catalysts. The authors conclude that Jemplate can help users better understand and utilize LLMs in the design process, fostering deeper and more reflective ideation.The paper "Jemplate: Exploring LLM-Enhanced Templates for Idea Reflection" by Xiaotong (Tone) Xu explores the integration of large language models (LLMs) into design templates to enhance idea reflection and critical thinking. The authors highlight the challenges of using chat-based LLMs, such as the need for users to craft effective prompts and the tendency for LLM responses to be lengthy and error-prone. To address these issues, they propose an alternative approach using structured templates, which are already effective tools for helping novices organize and reflect on information. The study includes formative research with 52 students and five instructors, revealing that novice designers often lack guidance on how to use LLMs effectively. Based on these insights, the authors developed Jemplate, a digital whiteboard plugin that integrates LLM capabilities into design templates. Jemplate streamlines the collection and organization of user-generated content and LLM responses within the template structure, providing reflective questions and in-situ guidance to help users think critically and improve their ideas. A preliminary study with 8 novice designers showed that Jemplate's reflective questions and in-situ guidance improved their ability to think critically and enhance their ideas. The paper discusses the potential of LLM-enhanced templates to instigate critical reflection and provides design guidelines for integrating LLMs into templates, including leveraging template integration, precision prompting, and reflection catalysts. The authors conclude that Jemplate can help users better understand and utilize LLMs in the design process, fostering deeper and more reflective ideation.