Youheng Bai

Cognitive diagnosis aims to infer learners' knowledge states from their exercise responses, enabling personalised education at scale. Existing methods represent exercises solely by coarse‐grained knowledge component annotations, overlooking semantic content and step‐level cognitive processes. This causes diagnostic confounding, in which exercises requiring fundamentally different reasoning pathways become indistinguishable in learned representations. Such confounding degrades diagnostic accuracy by conflating distinct learning deficiencies. To address this, we propose H2CD, a semantic‐enhanced heterogeneous hypergraph framework that unifies LLM‐based cognitive decomposition with structural learning. First, we leverage a large language model to decompose exercises into step‐by‐step solution paths and extract fine‐grained step‐level KCs, capturing procedural knowledge beyond predefined annotations. Second, we construct a heterogeneous hypergraph with two types of hyperedges: knowledge hyperedges that connect exercises to their associated multi‐granular KCs, preserving compositional knowledge structures, and interaction hyperedges that capture personalised student‐exercise relationships. Third, a two‐stage hypergraph neural network performs knowledge structure propagation on static exercise‐KC relationships and interaction propagation on dynamic student‐exercise patterns. Finally, a KC‐aware prediction layer projects representations to a learnable diagnostic axis, enabling interpretable mastery assessment at individual KC granularity. Extensive experiments on three real‐world educational datasets demonstrate that H2CD outperforms state‐of‐the‐art methods in both AUC and accuracy.

Prerequisite relation (PR) learning is a fundamental task in educational technology that identifies dependencies between learning resources to facilitate personalized learning experiences and optimize educational content organization. This survey provides a systematic review of prerequisite relation learning, emphasizing both methodological advances and practical applications. We first explore two distinct granularities of learning resources: knowledge concepts (KCs) and learning objects (LOs), establishing their definitions and relationships. We then introduce a novel classification framework for prerequisite relation learning methods based on both feature types and enhancement relationships, categorizing existing approaches into four types: (1) multi-source knowledge features for KCs’ prerequisite relation learning; (2) semantic knowledge features for LOs’ prerequisite relation learning; (3) LOs-enhanced learning for KCs’ prerequisite relation learning; and (4) KCs-enhanced learning for LOs’ prerequisite relation learning. The survey highlights recent developments in modeling KCs’ prerequisite relations. We provide a comprehensive analysis of evaluation methodologies, including both intrinsic metrics and extrinsic evaluation. Furthermore, we analyze the practical impact of prerequisite relations in educational applications, from adaptive learning path generation to curriculum design. Finally, we discuss current challenges and future opportunities for prerequisite relation learning.

Knowledge tracing (KT) models students' knowledge states and predicts their future performance based on their historical interaction data. However, attention based KT models struggle to accurately capture diverse forgetting behaviors in ever-growing interaction sequences. First, existing models use uniform time decay matrices, conflating forgetting representations with problem relevance. Second, the fixed-length window prediction paradigm fails to model continuous forgetting processes in expanding sequences. To address these challenges, this paper introduces LefoKT, a unified architecture that enhances attention based KT models by incorporating proposed relative forgetting attention. LefoKT improves forgetting modeling through relative forgetting attention to decouple forgetting patterns from problem relevance. It also enhances attention based KT models' length extrapolation capability for capturing continuous forgetting processes in ever-growing interaction sequences. Extensive experimental results on three datasets validate the effectiveness of LefoKT.