Double-Edged Sword or Sharp Tool? Designing and Evaluating Triadic LLM-Teacher Collaboration for K-12 Writing at Scale

2026-05-28

AI Summary

This study addresses the challenges of large language model (LLM) misuse or inefficient deployment in K–12 writing instruction, coupled with heavy teacher workloads and inconsistent feedback quality. To tackle these issues, the authors develop a triadic collaboration framework grounded in systemic functional linguistics, strategically allocating roles such that LLMs generate initial drafting suggestions while teachers focus on quality assurance. The work introduces an innovative dynamic adaptive collaboration framework, uncovers a ceiling effect in linguistic expansion, and establishes a multidimensional evaluation system based on suggestion trajectory tracking. Validated through a large-scale dataset comprising 57,954 essays from 10,195 students across 120 schools over two years, the system demonstrates significant improvements in student writing quality and effectively alleviates teacher burnout.

Abstract

The double-edged sword of integrating Large Language Models (LLMs) requires an effective triadic collaboration mechanism among LLMs, teachers and students, especially for K-12 education. By developing a triadic collaboration system to support K-12 writing learning, a multidimensional evaluation framework grounded in Systemic Functional Linguistics and the suggestion trajectory tracing pipeline, this paper contributes a large-scale empirical dataset involving $57,954$ essays from $10,195$ students across $120$ schools over two years. Our findings confirm the efficacy of this system in improving writing quality through a strategic labor division: the LLM serves as a generative engine to mitigate teacher burnout, and the teacher acts as a pedagogical gatekeeper and bridge to guarantee feedback quality. While both LLM and teacher are critical for skill improvement, we uncover a ceiling effect where excessive linguistic expansion yields diminishing marginal utility. These suggest a dynamically adaptive LLM-teacher collaboration as student proficiency increases.

Problem

• Large Language Models
• K-12 education
• triadic collaboration
• writing instruction
• teacher burnout

Innovation

• triadic collaboration
• suggestion trajectory tracing
• Systemic Functional Linguistics
• LLM-teacher division of labor
• ceiling effect