This research examines the use of advanced Natural Language Processing (NLP) technology, specifically GPT-3.5/4 models, to enhance text-based learning for neurodiverse students within higher education. Recognizing that conventional pedagogical resources often fail to meet the distinct needs of learners with neurodevelopmental differences, our research explores the hypothesis that NLP-enabled personalization of academic content can facilitate effective, equitable, and engaging educational experiences for college students. We present AImpathizer, a tool designed to adapt academic content into representations more congruent with the cognitive styles of neurodiverse students. AImpathizer is designed using a user-centric approach, highlighting the challenges and needs of neurodiverse college students. AImpathizer aims to incorporate more accessibility modifications in comparison to traditional academic content. The outcomes of this research aim to provide empirical evidence supporting the integration of NLP tools into educational environments in synergy with Universal Design for Learning (UDL) framework, paving the way for more inclusive educational practices.

Recent advances in large language models (LLMs) have brought AI-based systems into higher education, raising critical concerns regarding ethical and pedagogical implications. As AI-based systems become widely integrated in educational settings, questions about these concerns become increasingly important, yet current literature on best practices is still developing. Our work aims to guide researchers and educators by exploring the challenges and opportunities of AI-based systems in computer science higher education. We conducted a PRISMA-inspired literature review (N = 45), applying qualitative thematic analysis. Our findings suggest AI can foster a judgment-free space for underrepresented groups, especially considering the often competitive and defensive climate of computer science education, yet AI overuse may simultaneously undermine students’ belief in their competencies. We examine student and educator perspectives, linguistic nuances, policy considerations, and next steps. We finalize the discussion through classroom recommendations that reject the perceived dichotomy between academic AI policy and teaching AI literacy.

In recent years an increasing number of autistic students enrolled in college, many choosing computer science (CS) and related majors. However, the retention and graduation rates for autistic students are considerably lower than for neurotypical students. Research investigating factors that influence autistic students' success in college is sparse. More importantly, there is a scarcity of research that examines assessment techniques used and factors that influence the success of autistic CS students. Here, we report findings based on a systematic literature review$(\mathrm{N}=44)$that aims to examine the experience of autistic CS students to identify factors that influence their success given current assessment techniques. Our findings indicate that traditional accommodations, which in general are difficult to access and lack personalization, fall short of addressing many of the challenges autistic students face. The absence of a common vocabulary and communication style (e.g., verbal, instructions, teaching materials) and differences in learning style and cognitive processing between autistic students and neurotypical instructors hinders autistic students' academic success. Executive functioning (EF) challenges impact both academic and social aspects. However, accommodations targeted at addressing challenges related to EF are overlooked. Finally, despite many autistic students' affinity for technological interventions, their incorporation in accommodation is almost non-existent. We propose a set of actionable guidelines that could aid many of the challenges autistic students experience in CS programs, including designing a curriculum inspired by universal design, designing personalized accommodations, and incorporating technological interventions as part of the accommodation process.