paper-with-me

홈 › Papers

FlatFormer: A Flat Transformer Knowledge Tracing Model Based on Cognitive Bias Injection

2025-12-07 · Xiao-li Xia, Hou-biao Li arxiv

Knowledge Tracing (KT) models face a critical `Performance-Complexity Trap'': capturing complex cognitive dynamics like learning sessions and memory decay typically requires deep hierarchical architectures, which incur prohibitive computational costs for real-time deployment. To resolve this, we propose FlatFormer, a streamlined architecture based on the novel design paradigm of `Information Injection over Structural Stacking.'' Unlike parameter-heavy hierarchical models, FlatFormer leverages a standard flat Transformer augmented with two lightweight injection mechanisms: (i) a hybrid input encoding strategy combining learnable session identifiers with fixed sinusoidal step embeddings; and (ii) a pre-computed power-law bias integrated directly into attention logits to explicitly model the forgetting curve. Extensive experiments on four large-scale datasets (e.g., EdNet, Junyi) show that FlatFormer achieves state-of-the-art performance. For example, on the EdNet dataset, compared to the strongest hierarchical baseline (HiTSKT), its absolute AUC increased by 8.3%, while using less than 15% of parameters, and inference speed was about three times faster. These results validate that high cognitive fidelity does not necessitate architectural complexity.

📄 PDF Abstract BibTeX arXiv:2512.06629

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge Tracing

Similar Papers 제목 키워드 기반

FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformer

2023-01-20 · CVPR 2023 1 · Zhijian Liu, Xinyu Yang, Haotian Tang, Shang Yang 외

Transformer, as an alternative to CNN, has been proven effective in many modalities (e.g., texts and images). For 3D point cloud transformers, existing efforts focus primarily on pushing their accuracy to the state-of-th…

Autonomous Driving

Dual-Flattening Transformers through Decomposed Row and Column Queries for Semantic Segmentation

2022-01-22 · Ying Wang, Chiuman Ho, Wenju Xu, Ziwei Xuan 외

It is critical to obtain high resolution features with long range dependency for dense prediction tasks such as semantic segmentation. To generate high-resolution output of size $H\times W$ from a low-resolution feature …

Semantic Segmentation

Tracing Knowledge Instead of Paterns: Stable Knowledge Tracing with Diagnostic Transformer

2023-04-30 · Proceedings of the ACM Web Conference 2023 4 · Yin, Yu and Dai, Le and Huang, Zhenya and Shen 외

Knowledge Tracing (KT) aims at tracing the evolution of the knowledge states along the learning process of a learner. It has become a crucial task for online learning systems to model the learning process of their users,…

Contrastive LearningDiagnosticKnowledge Tracing

Disentangled Knowledge Tracing for Alleviating Cognitive Bias

2025-03-04 · Yiyun Zhou, Zheqi Lv, Shengyu Zhang, Jingyuan Chen

In the realm of Intelligent Tutoring System (ITS), the accurate assessment of students' knowledge states through Knowledge Tracing (KT) is crucial for personalized learning. However, due to data bias, $\textit{i.e.}$, th…

Knowledge Tracing

Improving Question Embeddings with Cognitiv Representation Optimization for Knowledge Tracing

2025-04-05 · Lixiang Xu, Xianwei Ding, Xin Yuan, Zhanlong Wang 외

The Knowledge Tracing (KT) aims to track changes in students' knowledge status and predict their future answers based on their historical answer records. Current research on KT modeling focuses on predicting student' fut…

Knowledge Tracing