Date
Publisher
arXiv
Large Language Models (LLMs) have demonstrated remarkable capabilities in
mathematical problem-solving. However, the transition from providing answers to
generating high-quality educational questions presents significant challenges
that remain underexplored. To advance Educational Question Generation (EQG) and
facilitate LLMs in generating pedagogically valuable and educationally
effective questions, we introduce EQGBench, a comprehensive benchmark
specifically designed for evaluating LLMs' performance in Chinese EQG. EQGBench
establishes a five-dimensional evaluation framework supported by a dataset of
900 evaluation samples spanning three fundamental middle school disciplines:
mathematics, physics, and chemistry. The dataset incorporates user queries with
varying knowledge points, difficulty gradients, and question type
specifications to simulate realistic educational scenarios. Through systematic
evaluation of 46 mainstream large models, we reveal significant room for
development in generating questions that reflect educational value and foster
students' comprehensive abilities.
Study at a glance
- Question: How well can large language models generate high-quality educational questions across middle school subjects?
- Setting and data: 900 Chinese evaluation samples across middle school mathematics, physics, and chemistry; 46 large language models.
- Method: Systematic evaluation of 46 large language models using a five-dimensional framework.
Why it matters
Large language models can solve mathematical problems, but the transition from providing answers to generating high-quality educational questions presents significant challenges that remain underexplored. Addressing this gap can help models generate pedagogically valuable and educationally effective questions.
Key findings
- In Chinese middle school math, physics, and chemistry, 46 mainstream AI models showed significant room for development in generating educationally valuable questions.
- The evaluated AI models showed significant room for development in generating questions that foster students' comprehensive abilities.
- Although AI models demonstrate remarkable capabilities in solving math problems, transitioning to generating high-quality educational questions presents significant challenges.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Study design
