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

TL;DR

The paper tackles the challenge of sustaining coherent, personalized DSP tutoring over long interactions by reframing retrieval, memory, and control as a coupled cognitive-evolution process. It introduces CogEvo-Edu, a hierarchical system with a Cognitive Perception Layer (CPL), Knowledge Evolution Layer (KEL), and Meta-Control Layer (MCL) to evolve student models, knowledge bases, and teaching policies jointly. Through the DSP-EduBench benchmark and an LLM-as-Judge ensemble, CogEvo-Edu achieves substantial performance gains over static RAG and single-agent baselines, demonstrating improved factual accuracy, contextual relevance, memory coherence, personalization, knowledge guidance, and strategy flexibility. The work highlights the value of dynamic, value-driven knowledge management and multi-agent orchestration for long-horizon educational interactions with LLM tutors.

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.

Figures (3)

• Figure 1: CogEvo-Edu: Cognitive Evolution Educational Multi-Agent Collaborative System
• Figure 2: Evaluation pipeline based on LLM-as-a-Judge principles.
• Figure 3: Comparative Evaluation Results Visualization Radar Chart