CogEvo-Edu: Cognitive Evolution Educational Multi-Agent Collaborative System

2025-11-29

AI Summary

Existing STEM education dialogue systems predominantly rely on single large language models (LLMs) and static retrieval-augmented generation (RAG), limiting their ability to model students’ evolving cognitive states, dynamically manage heterogeneous knowledge bases, and adapt pedagogical strategies—especially in complex domains like digital signal processing (DSP). This paper proposes a multi-agent collaborative framework for STEM education, featuring a three-tier architecture: cognition-aware modeling, knowledge evolution, and meta-control. Key innovations include a dual-memory mechanism, spatiotemporal value-driven knowledge activation and compression, and an inner-outer loop协同 hyperparameter self-adaptation method, enabling joint co-evolution of student profiling, knowledge representation, and instructional decision-making. Integrating structured memory modeling, semantic compression, forgetting mechanisms, and an LLM-as-a-Judge ensemble evaluator, our approach achieves a comprehensive score of 9.23 on the custom DSP-EduBench benchmark—up from 5.32—significantly outperforming static RAG and single-agent baselines.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Multiagent Learning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Large language models (LLMs) are increasingly deployed as conversational tutors in STEM education, yet most systems still rely on a single LLM with a static retrieval-augmented generation (RAG) pipeline over course materials. This design struggles in complex domains such as digital signal processing (DSP), where tutors must maintain coherent long-term student models, manage heterogeneous knowledge bases, and adapt teaching strategies over extended interactions. We argue that retrieval, memory, and control should be treated as a coupled cognitive evolution process. We instantiate this view in CogEvo-Edu, a hierarchical educational multi-agent system comprising a Cognitive Perception Layer (CPL), a Knowledge Evolution Layer (KEL), and a Meta-Control Layer (MCL). CPL maintains dual memories and performs confidence-weighted consolidation to build structured, self-correcting student profiles under limited context. KEL assigns each knowledge chunk a spatiotemporal value that drives activation, semantic compression, and forgetting. MCL formulates tutoring as hierarchical sequential decision making, orchestrating specialized agents and jointly adapting CPL/KEL hyperparameters via a dual inner--outer loop. To evaluate CogEvo-Edu, we construct DSP-EduBench, a vertical benchmark for DSP tutoring with heterogeneous resources, simulated student profiles, and long-horizon interaction scripts. Using a three-model LLM-as-a-Judge ensemble, CogEvo-Edu raises the overall score from 5.32 to 9.23 and improves all six indicators over static RAG, simple memory, and a single-agent variant, demonstrating the value of jointly evolving student profiles, knowledge bases, and teaching policies.

Problem

• Develops a multi-agent system to replace static LLM tutors in STEM education
• Addresses challenges in maintaining coherent student models and adapting teaching strategies
• Enhances tutoring effectiveness through coupled cognitive evolution of memory and control

Innovation

• Hierarchical multi-agent system with three cognitive layers
• Dual memory consolidation for self-correcting student profiles
• Spatiotemporal knowledge activation with adaptive forgetting