Multi-Agent Interactive Classroom
MAIC
A new kind of learning environment, where AI teachers and learning companions make online learning interactive and personalized.
Visit MAIC ↗Read the JCST paper ↗Tsinghua University · THU-MAIC
Jifan Yu 于济凡
Assistant Professor, School of Education
Building AI that
helps people learn.
I study generative AI for education, pedagogical agents, and multi-agent learning environments. I lead the THU-MAIC team at Tsinghua University.
AI × EducationContact ↗
Featured work
Reimagining how we learn.
Open source · Built for everyone
OpenMAIC
The open-source project behind MAIC. Explore the code and build with multi-agent educational environments.
Explore on GitHub ↗CCF 2026 公益优秀案例 · 全国仅 10 个GitHub Trending #1 · Multiple times in 2026
MAIC-UIInteractive courseware with generative UI · UIST 2026MOOCCubeXA knowledge-centered educational repository
清小搭Tsinghua University's Top 10 News Stories of 2024
Research & background
About me
I am an Assistant Professor at the School of Education, Tsinghua University, where I lead the THU-MAIC Team. I received my Ph.D. in Computer Science and Technology from Tsinghua University in January 2024, supervised by Prof. Juanzi Li and Prof. Jie Tang, and my bachelor’s degree in Software Engineering from Beihang University in 2018. I was previously a Shuimu Scholar at Tsinghua University.
My research focuses on generative AI for education and the design of new learning environments. I develop knowledge-guided educational language models, safe and controllable pedagogical agents, multi-agent learning environments, and personalized learning algorithms. I also explore AI-assisted educational research, including learner modeling, learning behavior simulation, and automated collection, analysis, and assessment of educational data.
I am looking for self-motivated collaborators, research assistants, and prospective master’s students interested in AI for education.
Full curriculum vitae ↗Recent research
Selected publications
UIST 2026 (Accepted)
MAIC-UI: Making Interactive Courseware with Generative UI
About this paper
A system for generating and incrementally editing interactive courseware from textbooks, slides, and PDFs.
Journal of Computer Science and Technology (JCST), 41(1): 394–414
From MOOC to MAIC: Reimagine Online Teaching and Learning Through LLM-Driven Agents
About this paper
A multi-agent approach to interactive online teaching and learning, extending the MOOC model with LLM-driven educational agents.
ACL 2026
SimPBL: A Multi-Agent Framework for Project-Based Learning
ACL 2026
From Knowing to Teaching: Scaffolding Pedagogical Decisions for LLM Agent
Learning together
Meet the students
Our growing research community at Tsinghua University.
Grants, awards & invited talks
Selected Research Grants
- Principal Investigator, National Natural Science Foundation of China Young Scientists Fund, 2025–2027: Knowledge- and LLM-integrated intelligent guidance for online education (融合知识与大模型的在线教育智能导学关键技术研究).
- Principal Investigator, CIPS-SMP Interdisciplinary Large Model Fund, 2025–2026: Explainable student cognitive diagnosis with LLM workflows (基于大模型工作流的可解释学生认知诊断关键技术研究).
- Subproject Lead, Tsinghua University Guoqiang Institute, 2024–2026: A virtual campus system powered by LLM-based multi-agent technology (基于大模型多智能体的清华大学虚拟校园系统).
Selected Awards & Recognition
- 2026: OpenMAIC received the CCF Outstanding Public Welfare Case recognition (CCF 公益优秀案例), one of only 10 selected cases nationwide.
- 2026: Outstanding Case on AI Safety and AI for Good (人工智能安全向善优秀案例), Ministry of Education and Cyberspace Administration of China.
- 2025: First Prize, Wu Wenjun AI Science and Technology Progress Award (吴文俊人工智能科技进步一等奖), for our work on massive open online courses.
- 2025: Exhibition demonstration project at the World Digital Education Conference, “全AI守护课堂”.
- 2024: Selected for the Ministry of Education’s first batch of “AI + Higher Education” typical applications, one of 18 selected projects nationwide.
- 2024: Tsinghua University’s Top 10 News Stories, “清小搭”; Tsinghua Shuimu Scholar.
- 2023: EMNLP Outstanding Paper Award.
- 2023: ACL Best Demo Paper Award for VisKoP.
- 2021: CIKM Best Resource Paper Nomination for MOOCCubeX.
Selected Invited Talks
- 2026: Invited talk at UNESCO Headquarters.
- 2025: “当MCP赋能‘MCP’：多智能体数智学习模式”, China National Computer Congress (CNCC), Harbin, China.
- 2025: “Reshaping Online Learning via Emergence of LLM-driven Multiple Agents”, AICET Workshop, Singapore.
- 2024: “面向认知的大模型自主智能体技术及其教育应用”, World Artificial Intelligence Conference (WAIC), Shanghai, China.
- 2020: “The Knowledgeable Intelligence in MOOCs”, AI Time. Slides