Fine-grained prompting substantially improves question uniqueness and usability of LLM-generated educational questions compared with Chain-of-Thought prompting¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · no dated study recorded · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Fine-grained prompting raises usability scores across all models in the computer science task, with Qwen2.5-7B increasing from 74.56% to 98.94%. → fgp-uniqueness-2

Evidence¶

fgp-uniqueness-2¶

Xiaolong Wang, Zhe Zhao, Song Lai, Chaoli Zhang, Zijie Geng, Yu Tong, Ye Wei, and Qingsong Wen. 2026. From Memorization to Creation: Evaluating the Cognitive Depth of LLM-Generated Educational Questions. Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1 (KDD '26). https://doi.org/10.1145/3770854.3785686

q2 · i? · causal · r1

Usability analysis after filtering, reported in the RQ2 results section for the computer science task under FGP. Qwen2.5-7B's usability "increases from 74.56% to 98.94%", with GLM-4-9B and InternLM3-8B reaching 99.43% and 99.08%.

"In the computer science task, the usability score ofQwen2.5-7Bincreases from 74.56% to 98.94%, whileGLM-4-9BandInternLM3-8Breach 99.43% and 99.08%, respectively."

Discussion¶

Related Claims¶

• Fine-grained prompting improves Bloom-level category consistency and knowledge coverage of generated questions in the computer science domain — related
• The Bloom-driven framework generalizes to social-science questions, with automated rubrics agreeing strongly with experts on usability but only modestly on Bloom level — related
• Cognitive leap is the most prevalent deviation pattern of LLM-generated questions, and fine-grained prompting amplifies upward shifts while suppressing cognitive regression — related