Exportar Publicação

Botas, J., Peixoto, A. & Ribeiro, E. (2026). BERT-Based Classification of Cross-Linguistic MCQs According to the Bloom Taxonomy. In Batista, Fernando and Ribeiro, Eugénio and Ribeiro, Ricardo and Santos, André L. (Ed.), 15th Symposium on Languages, Applications and Technologies (SLATE 2026). (pp. 4:1-4:17).: Schloss Dagstuhl -- Leibniz-Zentrum für Informatik.

J. F. Botas et al.,  "BERT-Based Classification of Cross-Linguistic MCQs According to the Bloom Taxonomy", in 15th Symp. on Languages, Applications and Technologies (SLATE 2026), Batista, Fernando and Ribeiro, Eugénio and Ribeiro, Ricardo and Santos, André L., Ed., Schloss Dagstuhl -- Leibniz-Zentrum für Informatik, 2026, vol. 144, pp. 4:1-4:17

@inproceedings{botas2026_1787537548923,
	author = "Botas, J. and Peixoto, A. and Ribeiro, E.",
	title = "BERT-Based Classification of Cross-Linguistic MCQs According to the Bloom Taxonomy",
	booktitle = "15th Symposium on Languages, Applications and Technologies (SLATE 2026)",
	year = "2026",
	editor = "Batista, Fernando and Ribeiro, Eugénio and Ribeiro, Ricardo and Santos, André L.",
	volume = "144",
	number = "",
	series = "",
	doi = "10.4230/OASIcs.SLATE.2026.4",
	pages = "4:1-4:17",
	publisher = "Schloss Dagstuhl -- Leibniz-Zentrum für Informatik",
	address = "",
	organization = "",
	url = "https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.SLATE.2026.4"
}

TY  - CPAPER
TI  - BERT-Based Classification of Cross-Linguistic MCQs According to the Bloom Taxonomy
T2  - 15th Symposium on Languages, Applications and Technologies (SLATE 2026)
VL  - 144
AU  - Botas, J.
AU  - Peixoto, A.
AU  - Ribeiro, E.
PY  - 2026
SP  - 4:1-4:17
DO  - 10.4230/OASIcs.SLATE.2026.4
UR  - https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.SLATE.2026.4
AB  - The Bloom Taxonomy provides a useful framework for describing the cognitive demands of Multiple Choice Questions (MCQs), yet automatically assigning Bloom levels remains challenging due to category overlap and the limited evidence of higher-order thinking in MCQs. This study aims to evaluate transformer-based models, including BERT and ModernBERT variants, to enhance their performance on a four-level Bloom classification task across multiple scenarios, while also conducting model interpretability and error analysis. 
Across 40 settings tested, BERT models consistently outperform a keyword-based baseline, highlighting the contextual and semantic representations for this task. Performance is broadly similar between BERT and ModernBERT, with only minor differences across architectures, while the inclusion of answer options yields only small gains, suggesting that the question stem alone contains most of the discriminative signal. Besides, cross-linguistic experiments show comparable results between the original English data and Portuguese, although translated data exhibits a slight performance degradation, likely due to direct translation noise and subtle syntactic shifts. Errors are concentrated between adjacent Bloom levels, and models perform best on the Remembering level. In contrast, the Applying level remains the most difficult class to predict, largely because of class imbalance and conceptual overlap.
The findings indicate that our approach constitutes a stable pipeline for Bloom classification, but its performance remains constrained by the intrinsic ambiguity of the taxonomy and the structural characteristics of the available data.
ER  -