Agentic Workflow for Education: Concepts and Applications

2025-09-01

AI Summary

Educational systems suffer from insufficient personalization, excessive teacher workload, and rigid instructional workflows. Method: This paper proposes AWE (Agent-based Workflow for Education), a multi-agent system grounded in the von Neumann architecture, integrating self-reflection, tool invocation, hierarchical task planning, and multi-agent coordination to transcend the inherent linear interaction paradigm of large language models and enable dynamic, non-linear pedagogical execution. Contribution/Results: AWE pioneers the deep integration of a scalable multi-agent architecture with educational workflows, supporting flexible orchestration and real-time adaptation. Empirical evaluation on automated mathematics item generation demonstrates statistical equivalence (p > 0.05) between AWE-generated items and authentic examination questions across difficulty distribution, knowledge-point coverage, and item-type structure, validating both efficacy and practical applicability.

Technology Category

• Multiagent Systems: Teamwork
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

With the rapid advancement of Large Language Models (LLMs) and Artificial Intelligence (AI) agents, agentic workflows are showing transformative potential in education. This study introduces the Agentic Workflow for Education (AWE), a four-component model comprising self-reflection, tool invocation, task planning, and multi-agent collaboration. We distinguish AWE from traditional LLM-based linear interactions and propose a theoretical framework grounded in the von Neumann Multi-Agent System (MAS) architecture. Through a paradigm shift from static prompt-response systems to dynamic, nonlinear workflows, AWE enables scalable, personalized, and collaborative task execution. We further identify four core application domains: integrated learning environments, personalized AI-assisted learning, simulation-based experimentation, and data-driven decision-making. A case study on automated math test generation shows that AWE-generated items are statistically comparable to real exam questions, validating the model's effectiveness. AWE offers a promising path toward reducing teacher workload, enhancing instructional quality, and enabling broader educational innovation.

Problem

• Developing agentic workflows for scalable personalized education
• Shifting from static prompts to dynamic multi-agent collaboration
• Reducing teacher workload through automated educational task execution

Innovation

• Four-component model with self-reflection and tool invocation
• Dynamic nonlinear workflows based on von Neumann architecture
• Multi-agent collaboration enabling personalized scalable execution