🧠 AI Ed Wiki

📄 Research Article

Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System

● high · 2026-07-16llm, generative-ai, intelligent-tutoring, higher-ed, stem-education, formative-assessment, benchmark, knowledge-tracing, personalized-learning, rag

LEA (Learning Engagement Assistant) is an agentic AI tutoring system that couples course-specific retrieval-augmented generation (RAG) with structured Knowledge Tracing / Knowledge Component (KC) models across integrated Chat, Tutor, and Quiz modes. This paper reports the first real-student classroom deployment of LEA (n = 8, STEM course CMP511) and the first empirical test of its cross-course scalability, extending a prior simulation-only validation that used synthetic learner agents.

The study exposes a key gap between simulated evaluation and real classroom deployment: synthetic learners predicted engagement patterns that diverged from observed behaviour, arguing that simulation alone cannot anticipate all aspects of live use. A RAGAS-based scalability evaluation across 660 questions found Answer Relevancy (0.88-0.94) and Context Precision (0.88-0.90) stable across courses, while Faithfulness declined with curriculum distance from LEA's original course (0.69 to 0.50) — a preliminary signal that downstream components, not the orchestration layer, constrain course-agnostic tutoring.

The work sits within the broader literature on Intelligent Tutoring and Agentic Workflows Education, with implications for Higher Ed and STEM Education deployment, and connects to debates on Formative Assessment quality and the limits of automated Feedback Loop in Personalized Learning.

Connected Concepts

  • Knowledge Tracing
  • Student Experience
  • Intelligent Tutoring
  • Higher Ed
  • STEM Education
  • Formative Assessment
  • Feedback Loop
  • Personalized Learning

    • Connected Articles

  • Agentic Workflows Education

    • Citation

    Teri Rumble, Javad Zarrin, P. George Lovell, Ruth Falconer (2026). Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System. arXiv:2607.13370.