AISI 愛思 | Adaptive Interactive Study Interface

核心理念 Core Principles

學習科學 DNA Learning-Science DNA

平台設計特別關注 AI 在教育場域中的角色定位，確保其回饋具備教學法一致性、歷程導向與可解釋性。 We position AI carefully in classrooms—feedback stays pedagogically aligned, process-oriented, and explainable.

🧭

高層次思考 Higher-Order Thinking

以高層次思考作為每個任務的錨點。 Higher-order thinking as the anchor of every task.

🔎

探究導向學習 Inquiry-First Learning

以探究為先的工作流程，強調證據與主張。 Inquiry-first workflows that foreground evidence and claims.

🧠

自我調整學習 Self-Regulated Learning

元認知提示促進自我調整學習。 Metacognitive nudges that foster self-regulated learning.

🤝

AI 作為教學夥伴 AI as Teaching Partner

AI 引導學習，但不取代教師或學生思考。 AI guides without replacing teachers or student thinking.

平台特色與功能 Platform Features

設計、引導、分析 Design, Guide, Analyze

模組化任務建置、多模態回應、歷程導向回饋與角色權限管理，讓教師與學生即時串連。 Modular task building, multimodal responses, process-aligned feedback, and role-based controls keep teachers and learners in sync.

AI 輔助任務與作業設計 AI-Assisted Task Design

創作

教師可透過模組化介面或 AI 對話式任務畫布，快速建立結合文本、圖片、影片與互動題型的學習任務，並自訂 AI 回饋與評分規準。 Build multimodal tasks through modular editors or conversational canvases—combine text, images, video, interactive items, and tailor AI feedback prompts plus rubrics.

歷程導向的 AI 回饋機制 Process-Oriented AI Feedback

引導

逐步引導、任務導向、反思回饋等模式，讓學生在「作答 → AI 回饋 → 再作答」的循環中深化理解。 Stepwise guidance, task-directed cues, and reflective prompts keep students iterating through answer → AI feedback → revise.

課程與角色權限管理 Course & Role Management

治理

完整的課程與角色區分，涵蓋公告、作業繳交、歷史紀錄與學習表現檢視，同一任務可彈性轉換課程作業。 Clear roles and course governance—announcements, submissions, histories, and performance dashboards; reuse tasks as assignments with one switch.

多模態學習回應與 AI 分析 Multimodal Response & AI Analysis

解析

支援手寫、繪圖、圖片上傳與圖文混合回應，AI 可針對圖像與文本進行分析，提供符合學習目標的建議。 Accept handwriting, drawings, image uploads, and mixed media; AI analyzes visuals plus text to deliver goal-aligned feedback.

ThinkForge：AI 探究對話系統 ThinkForge: AI Inquiry System

探究

內建逐步引導、直接回應、科學探究、閱讀探究等模式，可整合對話資料夾與學習分析，支援長時程專題。 Modes for stepwise tutoring, direct reply, science inquiry, and reading inquiry; integrates conversation folders and analytics for long-form projects.

適用對象與場域 Target Audience

AISI 的適用場域 Where AISI Fits

• 國中、高中與高等教育課程 • Secondary through higher education courses
• 高層次思考任務、探究、跨領域學習 • Higher-order thinking tasks, inquiry, interdisciplinary learning
• 自主學習課程、探究與實作導向教學 • Self-directed learning, inquiry, and practice-driven instruction

願景 Vision

AISI.tw 的未來 The Future of AISI.tw

AISI.tw 致力於成為「以學習科學為基礎的教育 AI 平台」，在生成式 AI 時代中，協助教師設計更有深度的學習活動，並支持學生發展思考力、反思力與學習自主性，而非被動依賴 AI。 We aim to be the learning-science-first AI platform that empowers educators to build depth, while students gain agency—not dependency.

發表 Publications

學術發表 Publications

期刊論文 Journal Articles

同儕審查

Sui, C.-J., & Chang, C.-Y. (2026). Generative AI and education: Exploring the interplay between teachers' epistemic beliefs and self-regulated learning. Educational Technology & Society, 29(3). https://doi.org/10.30191/ETS.202607_29(3).RP03
Sui, C.-J., Chang, C.-Y., & Yen, M.-H. (2025). STEM-5E Socio-scientific argumentation with generative AI-driven scaffolding: Exploring the interplay between epistemic beliefs and learning outcomes. Journal of Science Education and Technology. https://doi.org/10.1007/s10956-025-10257-6
Sui, C.-J. (in press). Enhancing the accuracy of LLM-based automatic scoring: A study of zero-shot, few-shot, and self-consistency strategies in socio-scientific issues. Contemporary Journal of Science Education.

會議論文 Conference Articles

學術社群

Sumardani, D., Sui, C.-J., Sreedevi, P. S., Yen, M.-H., & Chang, C.-Y. (2025, February). AI-driven Feedback on Argumentation: Should we use AI as our scaffolding on Energy Issue? Paper presented at 2025 IEEE 8th Eurasian Conference on Educational Innovation (IEEE ECEI 2025), Bali, Indonesia.
Hsiao, S. Y., & Chang, C.-Y. (2025, February). Cultivating Pre-service Teachers to Design Critical Thinking Curriculum Using the DECODE Model. Paper presented at 2025 IEEE 8th Eurasian Conference on Educational Innovation (IEEE ECEI 2025), Bali, Indonesia.
Sumardani, D., Yen, M.-H., & Chang, C.-Y. (2024, December). Exploring GenAI Implementation in Science Learning. Paper presented at The 40th Annual International Conference on Science Education (ASET 2024), Kaohsiung, Taiwan.
Sui, C.-J., Gijlers, H., Chen, H.-C., Yen, M.-H., & Chang, C.-Y. (2024, December). Investigate effect of GenAI-supported scaffolding on socio-scientific argumentation: A case study in the Netherlands. Paper presented at The 40th Annual International Conference on Science Education (ASET 2024), Kaohsiung, Taiwan.

計畫團隊 Project Team

AISI 背後的人們 People Behind AISI

主持人 Principal Investigator

張俊彥 Chun-Yen Chang

計畫主持人暨共同創辦人 · 國立臺灣師範大學 Principal Investigator & Co-founder · National Taiwan Normal University

國家講座教授 National Chair Professor

主持人 Principal Investigator

隋奇融 Chi-Jung Sui

計畫主持人暨共同創辦人 · 國家教育研究院 Principal Investigator & Co-founder · National Academy for Educational Research

助理研究員 Assistant Research Fellow

計畫夥伴 Project Partners

顏妙璇 Miao-Hsuan Yen

國立臺灣師範大學 · 科學教育所副教授 Associate Professor · Graduate Institute of Science Education, National Taiwan Normal University

于清華 Martin Yu

技術長 Chief Technology Officer

林易辰 Yi-Chen Lin

國立臺灣師範大學博士候選人 Ph.D. Candidate, National Taiwan Normal University

蕭聖益 Sheng-Yi Hsiao

新北市土城國中教師 · 國立臺灣師範大學博士生 Teacher, Tucheng Junior High School · Ph.D. Student, National Taiwan Normal University

金建鈺 Jianyu Jin

溫州大學小學教育系副主任 Deputy Director, Department of Primary Education, Wenzhou University

Dadan Sumardani

國立臺灣師範大學博士生 Ph.D. Student, National Taiwan Normal University

Arip Nurahman

Institut Pendidikan Indonesia 講師 · 印尼教育大學博士生 Lecturer, Institut Pendidikan Indonesia · Ph.D. Student, Indonesian University of Education

Bakhrul Rizky Kurniawan

瑪瑯國立大學物理系講師 · 國立臺灣師範大學博士生 Lecturer, Physics Department, State University of Malang · Ph.D. Student, National Taiwan Normal University

頁尾 Footer

本平台由國科會專題研究計畫經費支持。 This platform is supported by NSTC research project funding.

我們準備好共同設計更深度的學習體驗。 Ready to co-design deeper learning experiences.

聯絡我們回到頂部