Building and Implementing Model of a Human-AI Teacher Training Symbiotic System in Multi-Agent Collaborative Environments

Abstract

Teaching training is vital for teacher development, yet traditional models face inefficiency and rigidity. This study constructs a Human-AI Teaching Training Symbiotic System through multi-agent collaboration, grounded in symbiosis theory by analyzing educational MAS evolution. The model integrates knowledge co-creation, evidence-based research, and symbiotic practice, and proposes a five-phase framework—co-sensing, co-discussing, co-verifying, co-creating, and co-implementing—to externalize tacit knowledge, synergize collective wisdom, and dynamize practical evidence. This Human-AI collaborative paradigm offers theoretical-practical innovation for digital intelligence education, enhancing both teacher expertise and instructional quality.