Inclusive classes supported by Artificial Intelligence and interdisciplinarity

Instituto de Ciências Exatas e Informática

Abstract

This paper reports a two-year university extension initiative to include people who are deaf or have severe hearing loss, carried out with students at the CMAEE Dr. Tarso de Coimbra. The work combined university extension, active learning methodologies (based on CEDEM-USP training), and supervised use of large language models (LLMs) to create self-instructional quizzes suited to this audience, whose literacy in written Portuguese is limited. Computer Science and Physiotherapy students from PUC Minas supported the activities, which included game therapy and hands-on crafts. The systems were validated both technically and pedagogically. They produced greater learning and engagement than lectures with Brazilian Sign Language (Libras) interpretation.