TutorCaseBench: A Case-Grounded Benchmark for Evaluating LLM Tutors in Supporting Students with Challenging Behaviors
TutorCaseBench is introduced, a benchmark that grounds AI-tutor evaluation in qualitative case studies from pedagogy, educational psychology, and special education, and reveals a gap between intervention recognition and context-specific adaptation, especially for open-weight models.
Y. Park, Chanyoung Park, Tae Soo Kim et al.
· 0 citations