Towards LLM-Empowered Knowledge Tracing via LLM-Student Hierarchical Behavior Alignment in Hyperbolic Space
#Knowledge Tracing#Large Language Model#Hyperbolic Space#Educational Technology#Cognitive States#AAAI 2026#L-HAKT
📌 Key Takeaways
Researchers developed L-HAKT to improve knowledge tracing in education
The method addresses limitations in capturing hierarchical cognitive states and individualized problem difficulty
The approach uses contrastive learning in hyperbolic space between synthetic and real data
The framework was validated on four real-world educational datasets
The research was accepted to AAAI 2026 conference
📖 Full Retelling
Researchers led by Xingcheng Fu and six collaborators introduced a new method called Large Language Model Hyperbolic Aligned Knowledge Tracing (L-HAKT) on February 26, 2026, through arXiv to address limitations in existing educational knowledge tracing systems that fail to capture the hierarchical evolution of cognitive states and individualized problem difficulty perception. Knowledge Tracing (KT) is an educational approach that diagnoses students' concept mastery through continuous monitoring of learning states, but current methods primarily focus on behavioral sequences based on student IDs or shallow textual features, missing important nuances in how students learn and retain knowledge. The proposed L-HAKT framework utilizes a two-agent system where a teacher agent deeply parses question semantics and explicitly constructs hierarchical dependencies of knowledge points, while a student agent simulates learning behaviors to generate synthetic training data. The researchers then perform contrastive learning between synthetic and real data in hyperbolic space to reduce distribution differences in key features such as question difficulty and forgetting patterns. By optimizing hyperbolic curvature, the method explicitly models the tree-like hierarchical structure of knowledge points, precisely characterizing differences in learning curve morphology for knowledge points at different hierarchical levels. Extensive experiments conducted on four real-world educational datasets validated the effectiveness of this innovative approach, with the research paper accepted to the AAAI 2026 conference, highlighting its significance in advancing educational technology through artificial intelligence.
In mathematics, hyperbolic space of dimension n is the unique simply connected, n-dimensional Riemannian manifold of constant negative sectional curvature, often taken to be −1 for simplicity. It is homogeneous, and satisfies the stronger property of being a symmetric space. There are many ways to ...
Use of technology in education to enhance learning and teaching
Educational technology (commonly abbreviated as edutech or edtech) refers to the use of computer hardware, software, and educational theory and practice to facilitate learning and teaching. When referred to with its abbreviation, "EdTech", it often refers to the industry of companies that create edu...
A large language model (LLM) is a language model trained with self-supervised machine learning on a vast amount of text, designed for natural language processing tasks, especially language generation. The largest and most capable LLMs are generative pre-trained transformers (GPTs) that provide the c...
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2602.22879 [Submitted on 26 Feb 2026] Title: Towards LLM-Empowered Knowledge Tracing via LLM-Student Hierarchical Behavior Alignment in Hyperbolic Space Authors: Xingcheng Fu , Shengpeng Wang , Yisen Gao , Xianxian Li , Chunpei Li , Qingyun Sun , Dongran Yu View a PDF of the paper titled Towards LLM-Empowered Knowledge Tracing via LLM-Student Hierarchical Behavior Alignment in Hyperbolic Space, by Xingcheng Fu and 6 other authors View PDF HTML Abstract: Knowledge Tracing diagnoses students' concept mastery through continuous learning state monitoring in this http URL methods primarily focus on studying behavioral sequences based on ID or textual this http URL existing methods rely on ID-based sequences or shallow textual features, they often fail to capture (1) the hierarchical evolution of cognitive states and (2) individualized problem difficulty perception due to limited semantic modeling. Therefore, this paper proposes a Large Language Model Hyperbolic Aligned Knowledge Tracing(L-HAKT). First, the teacher agent deeply parses question semantics and explicitly constructs hierarchical dependencies of knowledge points; the student agent simulates learning behaviors to generate synthetic data. Then, contrastive learning is performed between synthetic and real data in hyperbolic space to reduce distribution differences in key features such as question difficulty and forgetting patterns. Finally, by optimizing hyperbolic curvature, we explicitly model the tree-like hierarchical structure of knowledge points, precisely characterizing differences in learning curve morphology for knowledge points at different levels. Extensive experiments on four real-world educational datasets validate the effectiveness of our Large Language Model Hyperbolic Aligned Knowledge Tracing (L-HAKT) framework. Comments: 9 pages, 6 figures, Accepted to AAAI 2026 Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2602.22879 [cs.AI] (or ar...
