HomeAbstract of 9th-globaletFostering Autonomous Learning in EFL Writing: An AI Translation Workflow with Proficiency-Tailored Prompt Disclosure

Fostering Autonomous Learning in EFL Writing: An AI Translation Workflow with Proficiency-Tailored Prompt Disclosure

  • Oct 20, 2026
  • Posted by:
  • Category: Abstract of 9th-globalet
No Comments


Abstract Book of the 9th Global Conference on Education and Teaching

Year: 2026

[PDF]

Fostering Autonomous Learning in EFL Writing: An AI Translation Workflow with Proficiency-Tailored Prompt Disclosure

Prof. Dr. Takeshi Okada

ABSTRACT:

This study aims to develop a novel translation workflow that iteratively integrates AI to foster autonomous learning among Japanese EFL writers at the elementary-to-intermediate proficiency levels. When assigned to write English journals, Japanese EFL students tend to use AI translation tools in two limited ways: (1) feeding raw Japanese text directly, and (2) relying passively on invisible automated preprocessing. In contrast, following our newly developed workflow encourages active self-discovery and independent reflection, helping learners improve their practical English writing and communication skills. While previous research has focused exclusively on linguistic and rhetorical differences between Japanese and English, this study incorporates specific learner profiles as parameters: grade levels, learning objectives in their textbooks, and English proficiency indicators such as TOEIC and TOEFL. We identify specific features of raw Japanese text that inevitably lower the quality of AI translation and have the AI itself generate a pre-editing prompt to convert the raw Japanese text, which serves as input for AI translation, into a logical, explicit format optimised for English translation. The AI then provides learners with helpful feedback and compelling reasons for corrections to their original Japanese text. Subsequently, the results of the direct and pre-processed translations are juxtaposed for comparative analysis.　This active reflection fosters the recognition of cross-linguistic differences and the acquisition of English-specific structural conventions. Ultimately, the proposed workflow elevates AI from a mere translation tool to a metacognitive learning assistant for EFL learners.

Keywords: Comparative analysis; Input optimization; Learner profiling; Metacognitive assistant; Pre-editing writing; template; sixth; edition; self-discipline