In-context examples improve LLM Bloom classification: three-shot and ten-shot prompting outperform zero-shot across datasets¶

Claim · All claims
Evidence · 1 study · causal r2 · 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? Both three-shot and ten-shot prompting outperform zero-shot prompting across datasets, though zero-shot still yields reasonable performance. → 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 · r2

RQ2 evaluation across five datasets and five prompting strategies (Table 3). Few-shot prompts with examples drawn from the same dataset beat zero-shot; zero-shot GPT-5 still achieved weighted F1 around 0.75 on most datasets.

"Both three-shot and ten-shot prompting outperform zero-shot prompting across datasets, suggesting that additional contextual information helps to align model predictions with Bloom's cognitive levels."

Discussion¶

Related Claims¶

• Supervised ML/DL Bloom question classifiers trained on one dataset drop substantially in weighted F1 when tested on unseen datasets — the general claim this finding bears on, with the studies behind it
• LLMs show robust zero- and few-shot Bloom classification of OOD AI-assisted questions compared with ML and BERT models — related
• Curricular chain-of-thought prompting improves competency-classification accuracy over zero-shot, with gains concentrated in larger models, while definition-based prompting does not improve performance — related
• The selected-questions-verbs prompting strategy yields the strongest LLM Bloom classification, reaching weighted F1 up to 0.84 — related
• Few-shot prompting does not improve diagnostic F1 or accuracy but reduces evidence over-attribution and false-attribution — reports the opposite
• Under Chain-of-Thought prompting, higher Bloom-level classification accuracy correlates with lower category consistency in generated questions — related
• Prompting with category information or few-shot examples yields limited improvement in distribution alignment, with zero-shot closest for most models — related
• Dual-Lens attention aligns with expert pedagogical focus, allocating most weight to Bloom's taxonomy verbs and problem-solving steps — related