Integrating AI for Enhanced Feedback in Translation Revision- A Mixed-Methods Investigation of Student Engagement
Project Overview
This document explores the application of generative AI, specifically ChatGPT, in the context of translation education by analyzing how Master's students utilize AI-generated feedback during their revision processes. Employing a mixed-methods approach, the study investigates the cognitive, affective, and behavioral dimensions of student engagement with the feedback. The findings indicate that students perceive the AI-generated feedback as beneficial for their learning; however, they also face difficulties in comprehending and effectively implementing the suggestions provided. This highlights a nuanced interaction between various engagement aspects, suggesting that while generative AI can enhance educational practices, it also necessitates further support to help students fully leverage its capabilities in their learning processes. Overall, the research underscores the potential of AI tools in education while pointing to the importance of addressing the challenges students encounter in integrating such technologies into their academic work.
Key Applications
ChatGPT-generated feedback on translation assignments
Context: Master's students in translation at a university in Hong Kong
Implementation: A specialized AI-powered Translation Teaching Platform was developed to integrate ChatGPT feedback into the translation revision process.
Outcomes: Students exhibited high levels of cognitive engagement and generally found the feedback easy to understand; however, they struggled with some meaning-level feedback. The integration of ChatGPT feedback is seen as a way to alleviate teacher workload and provide timely feedback.
Challenges: Students reported difficulties in comprehending meaning-level feedback and expressed a desire for more critical and specific suggestions.
Implementation Barriers
Cognitive barrier
Students struggled to understand some meaning-level feedback, which hindered their ability to effectively apply it.
Proposed Solutions: Provide clearer, more specific feedback and additional guidance on interpreting the feedback.
Engagement barrier
Some students showed skepticism towards the reliability of AI-generated feedback, leading them to rely on additional resources.
Proposed Solutions: Incorporate training on AI feedback usage and encourage critical evaluation of both AI and teacher feedback.
Project Team
Simin Xu
Researcher
Yanfang Su
Researcher
Kanglong Liu
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Simin Xu, Yanfang Su, Kanglong Liu
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai