The field of AI in education is witnessing significant developments, with a focus on enhancing the assessment process and providing personalized feedback to students. Researchers are exploring the potential of generative artificial intelligence (GenAI) and large language models (LLMs) to optimize grading, improve feedback quality, and promote deeper learning. Noteworthy papers in this area include The AI Tutor in Engineering Education, which proposes a transferable assessment protocol for STEM courses, and Hybrid Instructor Ai Assessment In Academic Projects, which demonstrates the effectiveness of a hybrid AI-instructor model in optimizing the assessment process. Additionally, papers like MATCH and Reasoning Path Divergence introduce innovative metrics and strategies for evaluating code and promoting diverse thinking in LLMs. Autograder+ is also a notable contribution, providing a multi-faceted AI framework for rich pedagogical feedback in programming education.
Advancements in AI-Driven Education and Code Evaluation
Sources
The AI Tutor in Engineering Education: Design, Results, and Redesign of an Experience in Hydrology at an Argentine University
Hybrid Instructor Ai Assessment In Academic Projects: Efficiency, Equity, And Methodological Lessons
MATCH: Task-Driven Code Evaluation through Contrastive Learning
Reasoning Path Divergence: A New Metric and Curation Strategy to Unlock LLM Diverse Thinking
Autograder+: A Multi-Faceted AI Framework for Rich Pedagogical Feedback in Programming Education