Action Research on Empowering College English Writing Classrooms with Generative Large Language Models ——From the Perspective of Integrating "Teaching, Learning, and Assessment"

Chen Ping

Chongqing Vocational and Technical University of Mechatronics

Abstract：

With the rapid development of generative large language models, it has become possible to apply such tools in college English writing instruction. However, how to reasonably integrate them into actual teaching practice and ultimately achieve the goal of integrating teaching, learning, and assessment requires further research. From the perspective of integrating teaching, learning, and assessment, this study adopts an action research approach by extensively utilizing generative large language models in a college English writing course over one semester. Centering on "promoting teaching through assessment and promoting learning through assessment," it explores teaching methods for leveraging large language models to support writing classrooms. The study proposes strategies for task design based on large language models, guidance methods for human-machine collaboration based on large language models, comprehensive real-time feedback mechanisms based on large language models, and data-driven instructional improvement approaches based on large language models. Practice has demonstrated that the application of large language models can effectively address the issues of delayed feedback and lack of specificity in traditional writing instruction, alleviate the disconnection between assessment and teaching to a certain extent, and advance the integration of teaching, learning, and assessment.

Key Words：

generative large language models; college english writing; integration of "Teaching, Learning, and Assessment"