Advancing Automatic Distractor Generation for English Vocabulary Tests Using Generative Artificial Intelligence
Maria Anastasia, Heru Chakra Setiawan, Asmarini, Teguh Bharata Adji, Indriana Hidayah · 2025
The rapid advancement of Artificial Intelligence (AI) technology has significantly influenced education, particularly through Automatic Item Generation (AIG). This research leverages Generative AI to create distractors for English vocabulary tests, aiming to produce high-quality and challenging distractors. The study explores the efficacy of Generative AI in generating effective distractors to increase the difficulty and reliability of tests, utilizing advanced AI models and comprehensive evaluations through expert and user assessments. The results indicate that Generative AI-generated distractors exhibit high accuracy and relevance, effectively deceiving test participants. The study highlights the potential of Generative AI to overcome limitations of traditional AIG methods by generating a wide variety of questions tailored to individual capabilities, enabling adaptive learning experiences and the rapid creation of extensive question banks. The implications suggest that educators can utilize Generative AI to enhance educational assessments. Future research should explore more advanced AI models and expand the application of Generative AI in generating diverse question types for tests like TOEFL and IELTS.