AI
CSTutorBench:ブロックベースプログラミングのチューターとしての小規模言語モデルのベンチマーキング
CSTutorBench: Benchmarking Small Language Models as Tutors for Block-Based Programming
arxiv2026年7月8日
日本語要約
CSTutorBenchは、ブロックベースプログラミングのチューターとして機能する小規模言語モデル(SLM)の効果を評価します。この研究は、AI駆動型教育の将来にとって重要であり、SLMが個別の学習体験と基本的なプログラミングスキルのサポートを提供する能力を示しています。
English Summary
CSTutorBench evaluates the effectiveness of small language models (SLMs) in acting as tutors for block-based programming. This research is significant for the future of AI-powered education, demonstrating SLMs' capability to provide personalized learning experiences and support for foundational programming skills.