COMA-LessonPlan

Introduction

A lesson plan (LP) is a structured guide outlining instructional objectives, methods, and assessments to ensure organized learning. However, existing LP creations are often time-consuming, inconsistent in structure, and lack pedagogical mechanisms for real-time adaptation to diverse learner needs. To address these issues, we propose a collaborative multi-role agent framework called COMA for automatic LP generation. COMA formulates LP generation as a collaborative workflow among multiple LLM agents with distinct pedagogical expertise: (1) the novice agent that represents a novice teacher possesses an overarching understanding of the intended lesson flow but demonstrates limited precision in implementing the specific instructional actions; (2) the veteran agent that represents an experienced teacher demonstrates deep familiarity with the curriculum, textbooks, and the knowledge components embedded in each unit; and (3) the master agent that represents a pedagogical expert exhibits a well-developed and confident grasp of lesson progression, with the ability to design, adapt, and implement specific instructional actions effectively and responsively. Through an iterative workflow, these agents collaboratively refine LP quality. Comprehensive experiments across five subjects, using expert-designed metrics, demonstrate that COMA significantly outperforms state-of-the-art methods, producing lesson plans with superior quality, coherence, and pedagogical alignment.

Demo & API

Online Demo

We provide an online demo for experiencing COMA's lesson plan generation capabilities:

• Demo Link: https://plan.100tal.com/home.html
• Password: Education

API Access

We offer COMA as a cloud-based API service. For API documentation and access requests, please contact:

📧 Email: zengxiaoli@stu.jnu.edu.cn

Running the Code

The results of baselines

The results generated by each baseline are saved in ./code/result/subject_results. Each .xlsx file represents a subject and stores the basic information of the results: Subject, Baseline, Id, Course Information, Lesson Plan Content.

Consistent

We use the $Kappa$ index as an indicator of scoring consistency calculation to verify the validity of the scoring. The scoring details of the baseline generation results are saved in ./code/result/consistent, which has five folders corresponding to five subjects. Each subject saves the baseline results independently, such as ChatGPT4prompt_chemistry. Each .xlsx file saves the results of three scorers. The first column is the ID, and the following are the scoring results of different dimensions. For the convenience of calculation, only the scoring results are retained. Finally, the consistency of each subject is saved in ./code/result/result_consistent.csv, the final result is saved in ./code/result/result_consistent.csv, and the scoring consistency details are saved in ./code/result.

Code

cd code
pip install -r requirements.txt
python consistent.py

Prompt

There are many prompt words used in COMA-LessonPlan. Some of them are now saved in ./code/prompt.