CSG: Cognitive Structure Generation for Intelligent Education

Hengnian Gu, Zhifu Chen, Yuxin Chen, Jin Peng Zhou, Dongdai Zhou
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:36960-36980, 2026.

Abstract

Cognitive structure (CS), a student’s construction of concepts and inter-concept relations, has long been recognized as a foundational notion in psychology and intelligent education, yet remains largely unassessable in practice. Existing approaches such as knowledge tracing (KT) and cognitive diagnosis (CD) simplify and indirectly approximate CS, but they intertwine representation learning with prediction objectives, limiting generalization, interpretability, and reuse across tasks. To address this gap, we propose Cognitive Structure Generation (CSG), a task-agnostic framework that explicitly models CS through generative modeling. Based on educational theories, CSG first pretrains a Cognitive Structure Diffusion Probabilistic Model (CSDPM) and then applies reinforcement learning with SOLO-based hierarchical rewards to capture plausible patterns of cognitive development. By decoupling cognitive structure representation from downstream prediction, CSG produces interpretable and transferable cognitive structures that can be seamlessly integrated into diverse student modeling tasks. Experiments on five real-world datasets show that CSG yields more comprehensive representations, substantially improving performance while offering enhanced interpretability and modularity.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-gu26b, title = {{CSG}: Cognitive Structure Generation for Intelligent Education}, author = {Gu, Hengnian and Chen, Zhifu and Chen, Yuxin and Zhou, Jin Peng and Zhou, Dongdai}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {36960--36980}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/gu26b/gu26b.pdf}, url = {https://proceedings.mlr.press/v306/gu26b.html}, abstract = {Cognitive structure (CS), a student’s construction of concepts and inter-concept relations, has long been recognized as a foundational notion in psychology and intelligent education, yet remains largely unassessable in practice. Existing approaches such as knowledge tracing (KT) and cognitive diagnosis (CD) simplify and indirectly approximate CS, but they intertwine representation learning with prediction objectives, limiting generalization, interpretability, and reuse across tasks. To address this gap, we propose Cognitive Structure Generation (CSG), a task-agnostic framework that explicitly models CS through generative modeling. Based on educational theories, CSG first pretrains a Cognitive Structure Diffusion Probabilistic Model (CSDPM) and then applies reinforcement learning with SOLO-based hierarchical rewards to capture plausible patterns of cognitive development. By decoupling cognitive structure representation from downstream prediction, CSG produces interpretable and transferable cognitive structures that can be seamlessly integrated into diverse student modeling tasks. Experiments on five real-world datasets show that CSG yields more comprehensive representations, substantially improving performance while offering enhanced interpretability and modularity.} }
Endnote
%0 Conference Paper %T CSG: Cognitive Structure Generation for Intelligent Education %A Hengnian Gu %A Zhifu Chen %A Yuxin Chen %A Jin Peng Zhou %A Dongdai Zhou %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-gu26b %I PMLR %P 36960--36980 %U https://proceedings.mlr.press/v306/gu26b.html %V 306 %X Cognitive structure (CS), a student’s construction of concepts and inter-concept relations, has long been recognized as a foundational notion in psychology and intelligent education, yet remains largely unassessable in practice. Existing approaches such as knowledge tracing (KT) and cognitive diagnosis (CD) simplify and indirectly approximate CS, but they intertwine representation learning with prediction objectives, limiting generalization, interpretability, and reuse across tasks. To address this gap, we propose Cognitive Structure Generation (CSG), a task-agnostic framework that explicitly models CS through generative modeling. Based on educational theories, CSG first pretrains a Cognitive Structure Diffusion Probabilistic Model (CSDPM) and then applies reinforcement learning with SOLO-based hierarchical rewards to capture plausible patterns of cognitive development. By decoupling cognitive structure representation from downstream prediction, CSG produces interpretable and transferable cognitive structures that can be seamlessly integrated into diverse student modeling tasks. Experiments on five real-world datasets show that CSG yields more comprehensive representations, substantially improving performance while offering enhanced interpretability and modularity.
APA
Gu, H., Chen, Z., Chen, Y., Zhou, J.P. & Zhou, D.. (2026). CSG: Cognitive Structure Generation for Intelligent Education. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:36960-36980 Available from https://proceedings.mlr.press/v306/gu26b.html.

Related Material