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

TL;DR

CogEvo-Edu introduces a hierarchical multi-agent system for STEM education that dynamically evolves student models, knowledge bases, and teaching strategies, outperforming static retrieval-based methods in complex digital signal processing tutoring.

Contribution

This paper presents CogEvo-Edu, a novel multi-layered system that integrates cognitive perception, knowledge evolution, and meta-control to enhance LLM-based tutoring in complex domains.

Findings

CogEvo-Edu significantly improves tutoring scores from 5.32 to 9.23.

The system outperforms static RAG and single-agent variants.

Joint evolution of student profiles, knowledge, and policies proves effective.

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.