Memory behavior modeling seeks to predict individual recall performance and understand its underlying cognitive mechanisms. However, the dynamic and heterogeneous nature of memory data poses significant challenges to the generalization ability of models under unseen conditions. To address this challenge, we propose an invariant representation learning framework I-Mem that integrates self-supervised contrastive learning with decorrelation constraints, enabling the adaptive identification and suppression of environment-related factors in sequential behavioral data, thereby mitigating the influence of spurious features and enhancing the modeling of stable cognitive structures. Importantly, the method does not rely on explicit environment partitioning or predefined environment labels, while our theoretical analysis demonstrates that it can effectively resist environmental perturbations and facilitate the extraction of invariant structural representations, thereby ensuring adaptability and generalization. Empirical evaluations on both synthetic and real-world datasets further confirm its superiority over mainstream methods in terms of generalization performance and stable feature identification. Feature attribution analysis reveals that I-Mem extracts invariant features aligned with classical cognitive effects, and reflects short-term behavioral patterns that may indicate latent cognitive mechanisms beyond existing theories, highlighting both interpretability and discovery potential.

Large language models (LLMs) have shown excellent performance in natural language processing but struggle with mathematical reasoning. As the training mode gradually solidifies, researchers propose a data-centric concept of artificial intelligence, emphasizing the development of higher-quality data to empower LLMs. Existing studies construct synthetic data for mathematical reasoning by expanding public datasets, thereby performing supervised fine-tuning of LLMs. However, these methods mostly focus on quantity while neglecting quality. The challenging samples fail to receive adequate consideration during data synthesis process, resulting in high construction costs, low-quality density, and serious data homogenization. This paper proposes a multi-agent environment called Virtual ClassRoom (VCR), which leverages various agents driven by LLM to construct high-quality diversified synthetic data. Inspired by the "Cone of Experience" educational theory, VCR introduces three experience levels (direct, iconic, and symbolic) into data synthesis process by analogy with human learning. A user-friendly instruction set and role-playing system are carefully designed, enabling VCR to autonomously plan the scale of synthetic data. This system covers various educational scenarios, including lecture, discussion, problem design and problem-solving. The Adaboost idea embodied in the global iterative process further promotes steady performance improvement. Extensive experiments show that the synthetic data generated by VCR possess higher quality density and generalization capability, which can give LLMs superior mathematical reasoning performance with the same scale.

Memory behavior modeling aims to predict individual performance over time and uncover underlying cognitive mechanisms. However, existing approaches often struggle to balance predictive accuracy, domain generalization, and model interpretability. To address this, we propose ROKAN, a cognitively inspired and symbolically interpretable memory modeling framework. Based on the Multiscale Context Model, ROKAN formalizes the evolution of memory traces as a differentiable Ordinary Differential Equation system, implemented via Kolmogorov-Arnold Networks to derive human-readable symbolic expressions. To enhance generalization across heterogeneous learning domains, we design an Adaptive Domain-Aware loss function, which integrates Empirical Risk Minimization with Distributionally Robust Optimization through dynamic domain-aware weighting. Our experiments demonstrate that ROKAN significantly outperforms existing mainstream methods in both predictive accuracy and domain generalization. The symbolic expressions were found to exhibit formal consistency with classical memory theories, which lends support to the model's theoretical assumptions and empirical performance, and provides a new pathway toward theoretically grounded white-box memory modeling. Our code is available at https://github.com/hellowads/ROKAN.

As online learning platforms continue to expand, the integration of large language models (LLMs) into course recommendation systems has shown significant potential, leveraging their semantic reasoning capabilities and inherent world knowledge. Current approaches typically adopt a small-model retrieval followed by an LLM-based re-ranking strategy. However, these methods face notable limitations. They often do not fully utilize the built-in knowledge of LLMs to enhance the initial retrieval phase and lack sufficient adaptation to the unique characteristics of educational contexts, limiting their effectiveness in personalized course recommendations. In addition, issues such as hallucinations and inconsistencies in the decoding of LLM output pose challenges to the reliability of these systems.To address these challenges, we propose CourseLRec, an innovative course recommendation framework that combines the semantic reasoning capabilities of LLMs with the efficiency of sequence modeling techniques. CourseLRec introduces three key innovations: LLM-Embedding for data augmentation of course titles and types, combined with Course-Aware Fusion to dynamically balance user interaction patterns and course content semantics; an Education-CoT Prompt to effectively integrate domain knowledge with learners’ progressive learning trajectories; and LoRA fine-tuning alongside a Context-Enhanced Token-to-Course Mapping module to enhance computational efficiency and semantic modeling. Comprehensive experiments on MOOCCourse and MOOCCube datasets demonstrate that CourseLRec outperforms state-of-the-art models across multiple evaluation metrics, highlighting its effectiveness in course recommendation tasks.

Memory behavior modeling is a key topic in cognitive psychology and education. Traditional approaches use experimental data to build memory equations, but these models often lack precision and are debated in form. Recently, data-driven methods have improved predictive accuracy but struggle with interpretability, limiting cognitive insights. Although knowledge-informed neural networks have succeeded in fields like physics, their use in behavior modeling is still limited. This paper proposes a Self-evolving Psychology-informed Neural Network (SPsyINN), which leverages classical memory equations as knowledge modules to constrain neural network training. To address challenges such as the difficulty in quantifying descriptors and the limited interpretability of classical memory equations, a genetic symbolic regression algorithm is introduced to conduct evolutionary searches for more optimal expressions based on classical memory equations, enabling the mutual progress of the knowledge module and the neural network module. Specifically, the proposed approach combines genetic symbolic regression and neural networks in a parallel training framework, with a dynamic joint optimization loss function ensuring effective knowledge alignment between the two modules. Then, for addressing the training efficiency differences arising from the distinct optimization methods and computational hardware requirements of genetic algorithms and neural networks, an asynchronous interaction mechanism mediated by proxy data is developed to facilitate effective communication between modules and improve optimization efficiency. Finally, a denoising module is integrated into the neural network to enhance robustness against data noise and improve generalization performance. Experimental results on five large-scale real-world memory behavior demonstrate that SPsyINN outperforms state-of-the-art methods in predictive accuracy. Ablation studies confirm the model's co-evolution capability, improving accuracy while discovering more interpretable memory equations, showing its potential for psychological research. Our code is released at: https://github.com/JiaqiDijon/SPsyINN

The rapid advancement of multimedia technologies and their increasing integration in education have underscored the importance of multimedia learning. Knowledge Tracing (KT) plays a crucial role in enabling adaptive multimedia learning by continuously monitoring students' progress and forecasting their performance throughout the learning process. Question lies at the heart of the KT process, making its representation crucial for building efficient KT models. However, the sparsity and complexity of question data pose significant challenges for existing methods to capture the underlying features of questions, thereby affecting the accuracy of knowledge state predictions. To address this issue, this paper attempts to introduce the diffusion model to the KT field, proposing a novel knowledge tracing model, DiffuQKT. The model presents a diffusion-based generative approach for question representation and enhances the stability of knowledge states through contrastive learning. Specifically, DiffuQKT first constructs question representations based on their concepts, difficulty, and variations, and then, during the forward phase, progressively adds noise to the question representations, disrupting them into a Gaussian distribution. In the reverse phase, DiffuQKT gradually recovers the representations from noise, generating higher-quality question representations for knowledge tracing. Furthermore, to guide more meaningful question generation, we incorporate question concepts and difficulty as conditions during the denoising process. In addition, to improve the robustness of knowledge states against subtle variations in question representations, we employ contrastive learning to stabilize knowledge states across both original and denoised question representations. We conduct extensive experiments on four public datasets, comparing DiffuQKT with 15 baseline methods. The results demonstrate that DiffuQKT significantly outperforms existing models. Moreover, we find that the diffusion-based generative approach for question representation proposed in this paper has the ability to significantly improve the performance of baseline models. The code can be found at https://github.com/lilstrawberry/DiffuQKT.

Incorporating a critiquing component into recommender applications facilitates the enhancement of user perception. Typically, critique-able recommender systems adapt the model parameters and update the recommendation list in real-time through the analysis of user critiquing keyphrases in the inference phase. The current critiquing methods necessitate the designation of a dedicated recommendation model to estimate user relevance to the critiquing keyphrase during the training phase preceding the recommendations update. This paradigm restricts the applicable scenarios and reduces the potential for keyphrase exploitation. Furthermore, these approaches ignore the issue of catastrophic forgetting caused by continuous modification of model parameters in multi-step critiquing. Thus, we present a general Representative Items Sampling Framework for Critiquing on Knowledge Graph Recommendation (RISC) implemented as a plug-in, which offers a new paradigm for critiquing in mainstream recommendation scenarios. RISC leverages the knowledge graph to sample important representative items as a hinge to expand and convey information from user critiquing, indirectly estimating the relevance of the user to the critiquing keyphrase. Consequently, the necessity for specialized user-keyphrase correlation modules is eliminated with respect to a variety of knowledge graph recommendation models. Moreover, we propose a Weight Experience Replay (WER) approach based on KG to mitigate catastrophic forgetting by reinforcing the user's prior preferences during the inference phase. Our extensive experimental findings on three real-world datasets and three knowledge graph recommendation methods illustrate that RISC with WER can be effectively integrated into knowledge graph recommendation models to efficiently utilize user critiquing for refining recommendations and mitigate catastrophic forgetting.

Knowledge Tracing (KT) is the task of predicting students’ future performance based on their past interactions with educational resources. A key aspect of KT is representation learning, which aims to capture meaningful features from students’ learning behaviors to improve prediction performance. Recently, contrastive learning methods have shown great promise in representation learning. As a result, KT models based on contrastive learning have been introduced to enhance representation learning for KT. However, these models have posed several challenges. Firstly, most of these models adopt the contrastive learning approach used in other fields, which involves data augmentation followed by contrastive learning, yet effectively applying data augmentation in KT remains an open challenge. Secondly, these models typically apply contrastive learning to only one of the fundamental components of KT: questions, interactions, or knowledge states, thereby limiting their overall performance. To address these issues, this article proposes a Multi-level Contrastive learning model for Knowledge Tracing (MCKT). MCKT (The code can be found at https://github.com/lilstrawberry/MCKT .) does not rely on data augmentation strategies; instead, it deeply integrates domain knowledge and performs contrastive learning at three levels: questions, interactions, and knowledge states. Experimental results on four publicly available datasets, compared against a total of 20 state-of-the-art KT models, demonstrate that MCKT consistently outperforms other models. Subsequent experiments further validate the effectiveness of the multi-level contrastive learning approach.

Knowledge Tracing (KT) aims to predict students’ future performance on answering questions based on their historical exercise sequences. To alleviate the problem of data sparsity in KT, recent works have introduced auxiliary information to mine question similarity, resulting in the enhancement of question embeddings. Nonetheless, there remains a gap in developing an approach that effectively incorporates various forms of auxiliary information, including relational information (e.g., question–student , question–skill relation), relationship attributes (e.g., correctness indicating a student's performance on a question), and node attributes (e.g., student ability ). To tackle this challenge, the Similarity-enhanced Question Embedding (SimQE) method for KT is proposed, with its central feature being the utilization of weighted and attributed meta-paths for extracting question similarity. To capture multi-dimensional question similarity semantics by integrating multiple relations, various meta-paths are constructed for learning question embeddings separately. These embeddings, each encoding different similarity semantics, are then fused to serve the task of KT. To capture finer-grained similarity by leveraging the relationship attributes and node attributes on the meta-paths, the biased random walk algorithm is designed. In addition, the auxiliary node generation method is proposed to capture high-order question similarity. Finally, extensive experiments conducted on six datasets demonstrate that SimQE performs the best among 10 representative question embedding methods. Furthermore, SimQE proves to be more effective in alleviating the problem of data sparsity.

Recommendation systems are the primary means of information filtering in the current information age. They utilize user historical behaviors, personal preferences, and item attributes to predict user interests and deliver personalized recommendations. Existing research has predominantly focused on improving the accuracy of recommendation systems. However, false correlation and insufficient personalization are issues that have received limited attention. This paper introduces the Causal Feature-Enhanced Collaborative Filtering recommendation algorithm (CFECF) which addresses concerns that significantly impact the generalization capability of recommendation algorithms and user satisfaction. CFECF adopts a causal relationship perspective and introduces a latent outcome model framework to assess the true causal relationships between user and item features. This approach addresses the problem of false correlation in recommendation systems and enhances the model’s generalization capability. To improve personalization, user interaction histories are analyzed to derive a prior indicator based on statistical distribution, indicating user interest directions. The model’s training incorporates this indicator to enhance user satisfaction. Comprehensive experiments conducted on four real-world datasets affirm that the proposed CFECF markedly enhances recommendation performance when compared to state-of-the-art collaborative filtering methods, effectively addresses false correlation and insufficient personalization issues in recommendation systems.

Advances in multimedia technology and its widespread application in education have made multimedia learning increasingly important. Knowledge Tracing (KT) is the key technology for achieving adaptive multimedia learning, aiming to monitor the degree of knowledge acquisition and predict students' performance during the learning process. Current KT research is dedicated to enhancing the performance of KT problems by integrating the most advanced deep learning techniques. However, this has led to increasingly complex models, which reduce model usability and divert researchers' attention away from exploring the core issues of KT. This paper aims to tackle the fundamental challenges of KT tasks, including the knowledge state representation and the core architecture design, and investigate a novel KT model that is both simple and powerful. We have revisited the KT task and propose the ReKT model. First, taking inspiration from the decision-making process of human teachers, we model the knowledge state of students from three distinct perspectives: questions, concepts, and domains. Second, building upon human cognitive development models, such as constructivism, we have designed a Forget-Response-Update (FRU) framework to serve as the core architecture for the KT task. The FRU is composed of just two linear regression units, making it an extremely lightweight framework. Extensive comparisons were conducted with 22 state-of-the-art KT models on 7 publicly available datasets. The experimental results demonstrate that ReKT outperforms all the comparative methods in question-based KT tasks, and consistently achieves the best (in most cases) or near-best performance in concept-based KT tasks. Furthermore, in comparison to other KT core architectures like Transformers or LSTMs, the FRU achieves superior prediction performance with approximately only 38% computing resources. Through an exploration of the ReKT model that is both simple and powerful, is able to offer new insights to future KT research. The code can be found at https://github.com/lilstrawberry/ReKT.

Knowledge Tracing (KT) is vital for education, continuously monitoring students' knowledge states (mastery of knowledge) as they interact with online education materials. Despite significant advancements in deep learning-based KT models, existing approaches often struggle to strike the right balance in granularity, leading to either overly coarse or excessively fine tracing and representation of students' knowledge states, thereby limiting their performance. Additionally, achieving a high-performing model while ensuring interpretability presents a challenge. Therefore, in this paper, we propose a novel approach called Multiscale-state-based Interpretable Knowledge Tracing (MIKT). Specifically, MIKT traces students' knowledge states on two scales: a coarse-grained representation to trace students' domain knowledge state, and a fine-grained representation to monitor their conceptual knowledge state. Furthermore, the classical psychological measurement model, IRT (Item Response Theory), is introduced to explain the prediction process of MIKT, enhancing its interpretability without sacrificing performance. Additionally, we extended the Rasch representation method to effectively handle scenarios where questions are associated with multiple concepts, making it more applicable to real-world situations. We extensively compared MIKT with 20 state-of-the-art KT models on four widely-used public datasets. Experimental results demonstrate that MIKT outperforms other models while maintaining its interpretability. Moreover, experimental observations have revealed that our proposed extended Rasch representation method not only benefits MIKT but also significantly improves the performance of other KT baseline models. The code can be found at https://github.com/lilstrawberry/MIKT.

Knowledge tracing (KT) has become an increasingly relevant problem in intelligent education services. Deep learning-based knowledge tracing (DLKT) achieves superb performance in terms of prediction accuracy, but it lacks of explainability, which makes us hard to trust or understand models. The previous work on explaining DLKT was mainly based on gradients or attention scores, which is susceptible to spurious correlations, reducing the credibility of the explanation. To address this limitation, in this paper, we propose a causal explanation method based on the genetic algorithm (GA), named Genetic Causal Explainer(GCE), which constructs a causal framework to estimate the attribution of subsequence to the predictions of DLKT models, and a genetic coding system is designed. Further, A multi-strategy initialization method inspired by domain prior knowledge is proposed, and a global empirical matrix is introduced to capture the causal correlation knowledge during the search process across instances, and guiding the mutation operators. The GCE as a post hoc explanation method can generate explanation results without affecting model training, and can be applied to analyze different DLKT models. Experimental results demonstrate the GCE perform better than other explanation methods in terms of accuracy and readability in quantitative assessments. Meanwhile, the GCE also shows good application prospects in mining educational laws and comparing KT models.

Recently, online education has become popular. Many e-learning platforms have been launched with various intelligent services aimed at improving the learning efficiency and effectiveness of learners. Graphs are used to describe the pairwise relations between entities, and the node embedding technique is the foundation of many intelligent services, which have received increasing attention from researchers. However, the graph in the intelligent education scenario has three noteworthy properties, namely, heterogeneity, evolution, and lopsidedness, which makes it challenging to implement ecumenical node embedding methods on it. In this article, an autobalanced multitask node embedding model is proposed, named MNE, and applied to the interaction graph, settling a few actual tasks in intelligent education. More specifically, MNE builds two purpose-built self-supervised node embedding learning tasks for heterogeneous evolutive graphs. Edge-specific reconstruction tasks are built according to the semantic information and properties of the heterogeneous edges, and an evolutive weight regression task is designed, aiding the model to perceive the evolution of learners' implicit cognitive states. Then, both aleatoric and epistemic uncertainty quantification techniques are introduced, achieving both task- and node-level weight estimation and instructing subtask autobalancing. Experimental results on real-world datasets indicate that the proposed model outperforms the state-of-the-art graph embedding methods on two assessment tasks and demonstrates the validity of the proposed multitask framework and subtask balancing mechanism. Our implementations are available at https://github.com/ccnu-mathits/MNE4HEN.

Graph embedding (GE) aims to acquire low-dimensional node representations while maintaining the graph’s structural and semantic attributes. Intelligent tutoring systems (ITS) signify a noteworthy achievement in the fusion of AI and education. Utilizing GE to model ITS can elevate their performance in predictive and annotation tasks. Current GE techniques, whether applied to heterogeneous or dynamic graphs, struggle to efficiently model ITS data. The GEs within ITS should retain their semidynamic, independent, and smooth characteristics. This article introduces a heterogeneous evolution network (HEN) for illustrating entities and relations within an ITS. Additionally, we introduce a temporal extension graph neural network (TEGNN) to model both evolving and static nodes within the HEN. In the TEGNN framework, dynamic nodes are initially improved over time through temporal extension (TE), providing an accurate depiction of each learner’s implicit state at each time step. Subsequently, we propose a stochastic temporal pooling (STP) strategy to estimate the embedding sets of all evolving nodes. This effectively enhances model efficiency and usability. Following this, a heterogeneous aggregation network is devised to proficiently extract heterogeneous features from the HEN. This network employs both node-level and relation-level attention mechanisms to craft aggregated node features. To emphasize the superiority of TEGNN, we perform experiments on several real ITS datasets and show that our method significantly outperforms the state-of-the-art approaches. The experiments validate that TE serves as an efficient framework for modeling temporal information in GE, and STP not only accelerates the training process but also enhances the resultant accuracy.

Knowledge tracing (KT), which estimates and traces the degree of learners' mastery of concepts based on students' responses to learning resources, has become an increasingly relevant problem in intelligent education. The accuracy of predictions greatly depends on the quality of question representations. While contrastive learning has been commonly used to generate high-quality representations, the selection of positive and negative samples for knowledge tracing remains a challenge. To address this issue, we propose an adversarial bootstrapped question representation (ABQR) model, which can generate robust and high-quality question representations without requiring negative samples. Specifically, ABQR introduces the bootstrap self-supervised learning framework, which learns question representations from different views of the skill-informed question interaction graph and facilitates question representations between each view to predict one another, thereby circumventing the need for negative sample selection. Moreover, we propose a multi-objective multi-round feature adversarial graph augmentation method to obtain a higher-quality target view, while preserving the structural information of the original graph. ABQR is versatile and can be easily integrated with any base KT model as a plug-in to enhance the quality of question representation. Extensive experiments demonstrate that ABQR significantly improves the performance of the base KT model and outperforms state-of-the-art models. Ablation experiments confirm the effectiveness of each module of ABQR. The code is available at https://github.com/lilstrawberry/ABQR.

Automatic recommendation has become an increasingly relevant problem for industries, which allows users to discover items that match their tastes and enables the system to target items at the right users. Graph neural networks have attracted many researchers' attention and have become a useful tool for recommendation. However, these models face two major challenges, which are heterogeneous information aggregation and aggregation weight estimation. In this article, we propose a graph neural networks-based recommendation model, i.e., a separated graph neural recommendation (SGNR) model, which achieves high-quality performance. SGNR separates BINs in recommendation systems into two weighted homogeneous networks for users and items, respectively, resolving the heterogeneous information aggregation problem. In addition, a propagation coefficient estimation method is proposed, which combines parametric and nonparametric estimation strategies. And, it is constructed with three characteristics, which are collaborative, side-information constrained, and adaptive. Thereinto, a three-hierarchy attention operator is contained for feature fusion, which optimizes the feature aggregation process via a more sensible and flexible propagation mechanism. Experimental results on four public databases indicate that the proposed methods perform better than the state-of-the-art recommendation algorithms on prediction accuracy in terms of quantitative assessments and achieve readability and interpretability to some extent.

Recommendation accuracy is a fundamental problem in the quality of the recommendation system. In this article, we propose an efficient deep matrix factorization (EDMF) with review feature learning for the industrial recommender system. Two characteristics in user’s review are revealed. First, interactivity between the user and the item, which can also be considered as the former’s scoring behavior on the latter, is exploited in a review. Second, the review is only a partial description of the user’s preferences for the item, which is revealed as the sparsity property. Specifically, in the first characteristic, EDMF extracts the interactive features of onefold review by convolutional neural networks with word-attention mechanism. Subsequently,${L}_{0}$norm is leveraged to constrain the review considering that the review information is a sparse feature, which is the second characteristic. Furthermore, the loss function is constructed by maximuma posterioriestimation theory, where the interactivity and sparsity property are converted as two prior probability functions. Finally, the alternative minimization algorithm is introduced to optimize the loss functions. Experimental results on several datasets demonstrate that the proposed methods, which show good industrial conversion application prospects, outperform the state-of-the-art methods in terms of effectiveness and efficiency.

Automatic recommendation has become an increasingly relevant problem to industries, which allows users to discover new items that match their tastes and enables the system to target items to the right users. In this article, we have proposed a deep learning based fully Bayesian treatment recommendation framework, DVMF, which has high-quality performance and ability to integrate any kinds of side information handily and efficiently. In DVMF, the variational inference technique and the reparameterization tricks are introduced to make DVMF possible to be optimized by the stochastic gradient-based methods, in addition, two novel deep neural networks have been constructed to infer the hyper-parameters of the distributions of latent factors from the knowledge of user and item, which are represented as low-dimensional real-valued vectors retaining primary features. Experimental results on five public databases indicate that the proposed method performs better than the state-of-the-art recommendation algorithms on prediction accuracy in terms of quantitative assessments.

Automatic recommendation has become an increasingly relevant problem to industries, which allows users to discover new items that match their tastes and enables the system to target items to the right users. In this paper, we propose a deep learning (DL) based collaborative filtering framework, namely, deep matrix factorization (DMF), which can integrate any kind of side information effectively and handily. In DMF, two feature transforming functions are built to directly generate latent factors of users and items from various input information. As for the implicit feedback that is commonly used as input of recommendation algorithms, implicit feedback embedding (IFE) is proposed. IFE converts the high-dimensional and sparse implicit feedback information into a low-dimensional real-valued vector retaining primary features. Using IFE could reduce the scale of model parameters conspicuously and increase model training efficiency. Experimental results on five public databases indicate that the proposed method performs better than the state-of-the-art DL-based recommendation algorithms on both accuracy and training efficiency in terms of quantitative assessments.