|
Title:
|
HUMAN PERCEPTIONS ON AI-GENERATED IMAGE AND AI-ASSISTIVE SENTENCE CORRECTIONS: ANALYZING LEARNERS CONTEXTUAL VOCABULARY LEARNING BEHAVIOURS WITH AN AI CHATBOT |
|
Author(s):
|
Kensuke Akao, Mohammad Nehal Hasnine, Mirai Yamada and Hiroshi Ueda |
|
ISBN:
|
978-989-8704-72-6 |
|
Editors:
|
Demetrios G. Sampson, Dirk Ifenthaler and Pedro Isaías |
|
Year:
|
2025 |
|
Edition:
|
Single |
|
Keywords:
|
Generative Artificial Intelligence (GenAI), Vocabulary, AI-Assisted Vocabulary Learning, Contextual Vocabulary Learning, Image Generation, Sentence Constru |
|
Type:
|
Full Paper |
|
First Page:
|
103 |
|
Last Page:
|
110 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
if you are a member please login 
|
|
Paper Abstract:
|
In recent years, the rapid technological development and widespread adoption of chatbots equipped with generative artificial intelligence (GenAI) based on Large Language Models (LLMs) have brought dramatic changes in the field of education. In e-learning, flexible progression and real-time feedback enabled by human-like interaction with artificial intelligence (AI), similar to a tutor, along with a shift in the nature of learning contents from 'learning with pre-prepared materials' to 'learning with generated materials', significantly expands the possibilities for personalized learning implementation. However, how accurately can contents generated by an AI model for general-purpose cases that have not been trained explicitly for educational purposes meet the quality in foreign language learning? is yet to be explored. In this study, native speakers and foreign language teachers validated AI-suggested corrections to learners' constructed sentences and images representing their learning context, which were generated by a GenAI-based contextual vocabulary learning application called Wordhyve-Glc during the learning of the Spanish and French languages. Their feedback clarified that the AI is to some extent capable of correcting sentences and generating images to represent learning contexts but also generates many undesirable ones. This paper discusses the current technical limitations of AI-assisted sentence correction and image generation aspects in foreign language learning by examining failure cases and identifying the characteristics of these issues. It also outlines the necessary steps for further advancement in the use of GenAI within the educational field. |
|
|
|
|
|
|
