How LLMs Support EFL Writing:A Case Study of K-12 English Learning Based on the EDIPT Model
Large Language Model (LLMs), the most stylish pre-trained language-agency model, has recently garnered unprecedented global attention. However, the rapid growth of LLMs’ exceptional performance in understanding and comprehending human language and completing various tasks conversationally has sparked debates regarding its educational implications and applications. Notably, it serves as a language consultant, companion, and assessment expert in Chinese English education, utilizing suggestions to enhance the quality of listening, speaking, reading, and writing instruction. This study is one of the initial efforts to explore the potential contribution of LLMs to the instruction and acquisition of writing English as a Foreign Language (EFL) in K-12. As a result, the study proposes a response strategy framework that combines the EDIPT model to promote and support future English learning development in K-12 education. Additionally, applying an LLMs can improve learning efficiency, reduce the teacher’s burden, and support large-scale personalized learning, the other side also brings challenges such as weakened learning motivation, less reliability of knowledge, and privacy and security issues. To enhance the writing abilities of EFL students for K-12, it is recommended that educators work alongside LLMs to provide feedback on students' writing in a generating.
Authors
- Xiaoshu Xu (ORCID: https://orcid.org/0000-0002-0667-4511)
- Ming Zhang (ORCID: https://orcid.org/0000-0003-2728-128X)
- Yunfeng Zhang (ORCID: https://orcid.org/0000-0001-6928-1742)
- Yi Dai (ORCID: https://orcid.org/0000-0001-6616-1892)
- Qianhui Panghe
Institutions
- Stamford International University (TH)
- Guangdong Polytechnic Normal University (CN)
- Macao Polytechnic University (MO)
- City University of Macau (MO)
Publication Details
- Published
- 2024-11-06
- DOI
- https://doi.org/10.1109/ieir62538.2024.10959858
- Primary Topic
- Arabic Language Education Studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00