CSTutorBench: A New Benchmark for Evaluating Small AI Tutors in K-12 Coding Education
Researchers introduced CSTutorBench, a benchmark designed to evaluate small language models (SLMs) as tutors for block-based programming in K-12 education. The benchmark focuses on VEX VR, a block-based robotics environment, and aims to help schools select affordable, private AI tutoring tools without relying on expensive proprietary systems.
Researchers have introduced CSTutorBench, a new benchmark for evaluating small language models (SLMs) as AI tutors in block-based programming education. The benchmark specifically targets VEX VR, a block-based robotics environment used in K-12 classrooms. The motivation behind CSTutorBench is that while large language models are increasingly explored as AI tutors, their deployment in K-12 settings raises concerns around privacy, cost, and reliance on proprietary models. Small language models offer a promising alternative, but selecting the right model for a specific educational context remains difficult—especially when the target domain, such as block-based programming, is largely absent from model training data. CSTutorBench provides a standardized way to assess how well different SLMs can tutor students in this specialized domain, helping educators and developers identify the most effective and practical options for classroom use.