TutorBench: A Benchmark for Assessing LLM Tutoring Capabilities

TutorBench is a challenging benchmark designed to assess the tutoring capabilities of large language models. It consists of examples drawn from three common tutoring tasks: generating adaptive explanations, providing actionable feedback, and promoting active learning through hint generation. The dataset was created by ScaleAI and last updated on Hugging Face in October 2025.

Use Cases

• Benchmarking LLM performance on generating adaptive explanations based on student confusion.
• Evaluating model capabilities for providing actionable feedback on student work.
• Assessing the effectiveness of LLMs in promoting active learning through hint generation.

Strengths

• Focuses on three distinct and challenging tutoring tasks.
• Created by ScaleAI, a known organization in AI data.
• Last updated on 2025-10-08, indicating recent maintenance.

Limitations

• Row count, column definitions, and sample data are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

ScaleAI

Collection Method

Likely curated or synthesized for benchmarking purposes.

Time Range

Unknown

Freshness

Last updated 2025-10-08 00:15:46

Geography

Unknown