K-12EduBench: A Benchmark for Evaluating Large Language Models’ Knowledge, Problem-Solving, and Educational Goal Cognition in K-12 Education
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (40) · 34459-34466
DOI: 10.1609/aaai.v40i40.40744 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yuqing Ye und 6 weiteren Autor:innen.
Abstract (Originalsprache)
Large language models hold great promise for transforming K-12 education, but there is an urgent need for systematic evaluation of their core educational capabilities. Existing benchmarks often overlook educational goal cognition and overemphasize answer accuracy, thereby failing to capture deeper subject-level knowledge ability and problem-solving ability. To address this gap, we introduce K-12EduBench: a benchmark for evaluating LLMs’ subject-level knowledge ability, subject-specific problem-solving ability, and educational goal cognition ability in K-12 education. K-12EduBench comprises four components: (1) a dataset of 2,640 objective and 619 subjective questions across nine subjects, annotated with answers, problem-solving processes, and cognitive-level labels; (2) nine Item Response Theory (IRT) models for estimating subject-level knowledge ability; (3) evaluation methods and metrics for assessing multi-step problem-solving ability; and (4) prompts and scoring rubrics for measuring alignment with target cognitive levels. Experiments on advanced LLMs show that education-optimized models consistently outperform general-purpose ones across all three abilities, while under-scaled models lag substantially. We observe a strong positive correlation between subject-level knowledge ability and subject-specific problem-solving ability. Despite gains in educational goal cognition ability, current models—even those tailored for education—still fall short of real-world instructional needs.
Zitieren
APA
Ye, Y., Zhou, X., Chen, Z., Li, D., Gu, H., Zhou, J. P., & Zhou, D. (2026). K-12EduBench: A Benchmark for Evaluating Large Language Models’ Knowledge, Problem-Solving, and Educational Goal Cognition in K-12 Education. Proceedings of the AAAI Conference on Artificial Intelligence, 40(40), 34459-34466. https://doi.org/10.1609/aaai.v40i40.40744
Vancouver
Ye Y, Zhou X, Chen Z, Li D, Gu H, Zhou JP, et al. K-12EduBench: A Benchmark for Evaluating Large Language Models’ Knowledge, Problem-Solving, and Educational Goal Cognition in K-12 Education. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(40):34459-34466. doi:10.1609/aaai.v40i40.40744
BibTeX
@article{Ye2026,
author = {Ye, Yuqing and Zhou, Xuan and Chen, Zhifu and Li, Dandan and Gu, Hengnian and Zhou, Jin Peng and Zhou, Dongdai},
title = {K-12EduBench: A Benchmark for Evaluating Large Language Models’ Knowledge, Problem-Solving, and Educational Goal Cognition in K-12 Education},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {40},
pages = {34459--34466},
doi = {10.1609/aaai.v40i40.40744}
}
EUVIMED