The selected-questions-verbs prompting strategy yields the strongest LLM Bloom classification, reaching weighted F1 up to 0.84¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Combining manually selected in-context examples with expert-curated level-specific action verbs achieved the highest weighted F1 scores across datasets. → Abdolali Faraji 2026

Evidence¶

Abdolali Faraji 2026¶

Abdolali Faraji, Mohammadreza Molavi, Zohreh Rasoulkhani, Mohammadreza Tavakoli, and Gábor Kismihók. (2026). Cross-Dataset Bloom Question Classification: Supervised Models and Prompted LLMs. https://arxiv.org/abs/2606.13684 doi:10.48550/arxiv.2606.13684

q2 · i? · causal · r1

LLM evaluation (RQ2, Table 3): GPT-5 and GPT-5-mini classified all questions in five datasets under five prompting strategies; the "selected-questions-verbs" prompt reached weighted F1 up to 0.84 on the Yahya dataset.

"Specifically, theselected-questions-verbsprompt achieves the strongest performance, reaching weighted F1-scores of up to 0.84 on the Yahya dataset."

Discussion¶

Related Claims¶

• In-context examples improve LLM Bloom classification: three-shot and ten-shot prompting outperform zero-shot across datasets — related
• LLMs show robust zero- and few-shot Bloom classification of OOD AI-assisted questions compared with ML and BERT models — related
• Dual-Lens attention aligns with expert pedagogical focus, allocating most weight to Bloom's taxonomy verbs and problem-solving steps — related