A Multi-agent Open Learning Platform for LLM Era's Online Education
中文 | English
Join our 💬 WeChat | View MAIC 📖 Technical Report
MAIC (Massive AI-empowered Courses) aims to create a highly immersive and configurable AI-powered learning classroom, built on large language models and incorporating multiple intelligent agents such as teachers, teaching assistants, and peers. This system is designed to assist students throughout the entire learning process, support teachers in intelligent lesson preparation, and enable high-level personalized learning for students, ultimately realizing an "autonomous classroom" experience where each student has their own tailored learning journey.
Conceptual Design: This project’s conceptual design encompasses both technological and educational philosophies:
- Technological Philosophy: The project aspires to construct an intelligent classroom environment fully empowered by multi-agent systems based on large language models. It aims to significantly enhance students' immersive learning experiences while further strengthening intelligent assistance for teachers during lesson preparation, instruction, and post-class activities, thus facilitating the iterative evolution of intelligent learning environments.
- Educational Philosophy: Leveraging the powerful generalization capabilities of large language models, this project seeks to simultaneously address the dual challenge of balancing "scalability" and "personalization" in the digitization of education. The project aims to lead a paradigm shift in digital education for the new era, comparable to the transformation brought by large-scale classrooms in the past.
- [2025.01.23] 🚀🚀🚀 Our Classroom Simulation Paper is accepted by NAACL 2025!
- [2024.12.16] 🚀🚀🚀 Our Paper of Silde2Lecture subsystem is accepted by KDD 2025!
- [2024.12.5] 📚📚📚MAIC's comprehensive documentation is available at https://doc.maic.chat/ (See below for Application)
For more Info about our latest research and team members, welcome to visit our project page: https://project.maic.chat/
A Brief Demo Video for MAIC: Watch the video
Demo Video for Academic Consultant: Watch the video
- This repository is released under the Apache-2.0 License.
- The models and platform of MAIC are completely free for academic research. after filling out a "questionnaire" for registration, you can cooperate with MAIC's source code and documentation.
As generate contents by learning a large amount of multimodal corpora, but they cannot comprehend, express personal opinions or make value judgement. Anything generated by MAIC agents does not represent the views and positions of the model developers
We will not be liable for any problems arising from the use of MAIC platform, including but not limited to data security issues, risk of public opinion, or any risks and problems arising from the misdirection, misuse, dissemination or misuse of the model.
👏 Welcome to explore key techniques of MAIC and other AI4Edu projects of our team:
MOOCCube | MOOC-Radar | XDAI
If you find our model/code/paper helpful, please consider cite our papers 📝 and star us ⭐️！
@article{yu2024mooc,
title={From mooc to maic: Reshaping online teaching and learning through llm-driven agents},
author={Yu, Jifan and Zhang, Zheyuan and Zhang-li, Daniel and Tu, Shangqing and Hao, Zhanxin and Li, Rui Miao and Li, Haoxuan and Wang, Yuanchun and Li, Hanming and Gong, Linlu and others},
journal={arXiv preprint arXiv:2409.03512},
year={2024}
}Our Teacher-side Paper:
@article{zhang2024awaking,
title={Awaking the Slides: A Tuning-free and Knowledge-regulated AI Tutoring System via Language Model Coordination},
author={Zhang-Li, Daniel and Zhang, Zheyuan and Yu, Jifan and Yin, Joy Lim Jia and Tu, Shangqing and Gong, Linlu and Wang, Haohua and Liu, Zhiyuan and Liu, Huiqin and Hou, Lei and others},
journal={arXiv preprint arXiv:2409.07372},
year={2024}
}Our Student-side Paper:
@article{zhang2024simulating,
title={Simulating classroom education with llm-empowered agents},
author={Zhang, Zheyuan and Zhang-Li, Daniel and Yu, Jifan and Gong, Linlu and Zhou, Jinchang and Liu, Zhiyuan and Hou, Lei and Li, Juanzi},
journal={arXiv preprint arXiv:2406.19226},
year={2024}
}
