EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

R 2 GCurL: Reinforced Robust Knowledge Tracing via Dynamic Graph Curriculum Learning

Tianhao Peng, Yanjun Pu, Yongqi Zhang, Yuchen Li, Jian Ren, Jie Luo, Kui Zhang, Haitao Yuan, Shuaiqiang Wang, Dawei Yin, Wenjun Wu

ACM Transactions on Information Systems · 31.07.2026 · 44 (6) · 1-32

DOI: 10.1145/3818614 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 23.07.2026. Verfasst von Tianhao Peng und 10 weiteren Autor:innen. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rise of AI in education, knowledge tracing (KT) has become important for modeling students’ knowledge from interaction data. However, existing methods still face three major challenges, including limited modeling of personalized exercise–concept relations, low robustness to noisy interactions, and inefficient training due to suboptimal data selection. To address these issues, we propose R 2 GCurL, a novel KT framework with two key designs. First, we recast KT as a graph classification problem and construct dynamic graphs from student responses, enabling the model to capture structural relations between exercises and concepts for more personalized KT. Second, we introduce a data-centric curriculum learning strategy based on dynamic graph entropy. Under our definition, pairwise dynamic graph entropy measures graph-transition continuity, where larger values indicate stronger structural similarity. Its sequence-level aggregation is used to derive a structure-aware difficulty signal for sample scheduling. On top of this, an RL-based scheduler further adapts batch selection based on model feedback and is especially beneficial under noisier and more unstable training regimes. Theoretical analysis shows that R 2 GCurL has lower computational complexity than existing graph-based KT models. Extensive experiments on five real-world datasets confirm its effectiveness, robustness, and generalizability, including as a plug-and-play enhancement for sequence-based KT models.

Zitieren

APA

Peng, T., Pu, Y., Zhang, Y., Li, Y., Ren, J., Luo, J., Zhang, K., Yuan, H., Wang, S., Yin, D., & Wu, W. (2026). R 2 GCurL: Reinforced Robust Knowledge Tracing via Dynamic Graph Curriculum Learning. ACM Transactions on Information Systems, 44(6), 1-32. https://doi.org/10.1145/3818614

Vancouver

Peng T, Pu Y, Zhang Y, Li Y, Ren J, Luo J, et al. R 2 GCurL: Reinforced Robust Knowledge Tracing via Dynamic Graph Curriculum Learning. ACM Transactions on Information Systems. 2026;44(6):1-32. doi:10.1145/3818614

BibTeX
@article{Peng2026,
  author = {Peng, Tianhao and Pu, Yanjun and Zhang, Yongqi and Li, Yuchen and Ren, Jian and Luo, Jie and Zhang, Kui and Yuan, Haitao and Wang, Shuaiqiang and Yin, Dawei and Wu, Wenjun},
  title = {R 2 GCurL: Reinforced Robust Knowledge Tracing via Dynamic Graph Curriculum Learning},
  journal = {ACM Transactions on Information Systems},
  year = {2026},
  volume = {44},
  number = {6},
  pages = {1--32},
  doi = {10.1145/3818614}
}
Kostenlos starten →