Visual query localization (VQL) is pivotal for constructing episodic memory from egocentric videos. However, current methods often rely on computationally intensive attention mechanisms and rigid one-shot regression, which inherently struggle to model uncertainty and exhibit limited adaptability to degraded query appearances. This contrasts sharply with the human brain’s selective encoding and iterative hypothesis verification processes for episodic memory. Inspired by the human brain’s ability to selectively filter irrelevant information and reconstruct vague memory fragments through generative inference, we propose BiOVQL, a brain-inspired one-stage VQL framework. First, inspired by the hippocampus’s selective retention mechanism, we propose the Hippocampus-like Query-guided Spatio-Temporal Compression (HQSTC) module. By leveraging a selective state space model, visual queries are treated as neuromodulators, dynamically gating the video stream to maintain a compact latent state. This guides the model to consistently focus on query-relevant visual cues, enabling query-conditioned feature compression and efficient spatio-temporal memory encoding. Second, inspired by the prefrontal cortex’s re-localization mechanisms, we propose the Prefrontal-like Generative Refinement Localization (PGRL) module. We leverage a diffusion model to reconstruct the localization process as iterative denoising from noise to certainty, which aligns well with the human visual system’s coarse-to-fine perceptual reasoning. This enhances the model’s robustness in handling spatial ambiguities and achieving precise spatio-temporal retrieval. We conducted extensive experiments on the Ego4D-VQ benchmark, demonstrating that BiOVQL achieves state-of-the-art performance with comparable computational efficiency, thus offering an efficient and brain-inspired paradigm for VQL.

Cognitive diagnosis aims to infer students' mastery levels over knowledge components from their learning interactions, supporting personalized education applications. However, existing models encode responses as binary correctness labels, discarding information about which specific option a student selected. This input-level information loss limits their ability to distinguish qualitatively different error types. As a result, providing interpretable diagnostic outputs becomes challenging. To address these limitations, we propose SACD, a semantic-augmented cognitive diagnosis framework that integrates LLM-based semantic analysis with student behavioral modeling. SACD comprises the following key components. First, an LLM-based exercise diagnostic generator analyzes exercise content and produces structured semantic annotations for each answer choice, capturing the specific misconceptions each option represents. Second, a kernel-based alignment mechanism projects semantic embeddings and behavioral representations into a unified kernel space, enabling effective fusion of heterogeneous information. Third, an interpretable diagnosis layer predicts student performance and generates fine-grained mastery estimates, which LLMs further process to produce actionable learning plans. Extensive experiments on three real-world datasets demonstrate that SACD achieves superior prediction accuracy while enabling interpretable, actionable diagnostics.

Knowledge tracing (KT) aims to personalize online education on large-scale web-based platforms by modeling students' evolving knowledge states from their interaction sequences. However, most KT models rely on a single encoder architecture (e.g., self-attention or RNN), with fixed inductive biases that fails to capture the diversity of learning behaviors. Specifically, student learning unfolds across multiple timescales, and interaction sequences contain diverse frequency components ranging from short-term variations to long-term trends. Our data-driven analysis reveals that existing encoders exhibit characteristic frequency biases (e.g., self-attention tends to emphasize low-frequency patterns), highlighting the limitations of any single architecture. To address this problem, we propose FA-KT, a frequency-aware mixture of heterogeneous experts framework. FA-KT combines self-attention, Mamba, CNN, and LSTM experts, each with complementary frequency biases. A frequency-aware router analyzes each sequence's frequency characteristics and adaptively combines experts to create dynamic, personalized encoders for individual students. Across five benchmark datasets, FA-KT consistently outperforms 20 strong KT baselines in predicting future performance. Code is available at https://pykt.org/.

Graph Transformers (GTs) are increasingly applied to social network analysis, yet their deployment is often constrained by fairness concerns. This issue is particularly critical in incomplete social networks, where sensitive attributes are frequently missing due to privacy and ethical restrictions. Existing solutions commonly generate these incomplete attributes, which may introduce additional biases and further compromise user privacy. To address this challenge, FairGE (Fair Graph Encoding) is introduced as a fairness-aware framework for GTs in incomplete social networks. Instead of generating sensitive attributes, FairGE encodes fairness directly through spectral graph theory. By leveraging the principal eigenvector to represent structural information and padding incomplete sensitive attributes with zeros to maintain independence, FairGE ensures fairness without data reconstruction. Theoretical analysis demonstrates that the method suppresses the influence of non-principal spectral components, thereby enhancing fairness. Extensive experiments on seven real-world social network datasets confirm that FairGE achieves at least a 16% improvement in both statistical parity and equality of opportunity compared with state-of-the-art baselines.

Semantic understanding of popularity bias is a crucial yet underexplored challenge in recommender systems, where popular items are often favored at the expense of niche content. Most existing debiasing methods treat the semantic understanding of popularity bias as a matter of diversity enhancement or long-tail coverage, neglecting the deeper semantic layer that embodies the causal origins of the bias itself. Consequently, such shallow interpretations limit both their debiasing effectiveness and recommendation accuracy. In this paper, we propose FairLRM, a novel framework that bridges the gap in the semantic understanding of popularity bias with Recommendation via Large Language Model (RecLLM). FairLRM decomposes popularity bias into item-side and user-side components, using structured instruction-based prompts to enhance the model's comprehension of both global item distributions and individual user preferences. Unlike traditional methods that rely on surface-level features such as ''diversity'' or ''debiasing'', FairLRM improves the model's ability to semantically interpret and address the underlying bias. Through empirical evaluation, we demonstrate that FairLRM enhances fairness and recommendation accuracy through a trustworthy, semantically grounded treatment of popularity bias. The source code is shown in https://github.com/LuoRenqiang/FairLRM.

Knowledge tracing (KT) is a problem of modeling students’ knowledge states to predict their future performance by observing their historical learning interactions. The collection of educational data presents significant challenges, as students’ limited learning engagement restricts the generation of large-scale interaction data, while stringent privacy regulations further limit the availability of student learning sequences from online platforms. Hence, it is crucial to enhance the capabilities of deep learning-based KT (DLKT) models by constructing large-scale datasets through the integration of student interaction data across multiple subjects and sources. The success of ChatGPT demonstrates that the decoder-only Transformer architecture is highly effective in capturing complex information from large-scale sequential data. Against this background, we propose a novel decoder-only Transformer architecture-based model, named Unified DLKT ( UniKT ), to learn coherent and unified representations across a wide range of data sources. Specifically, we combine student learning sequences from six educational scenarios and utilize a multi-source encoding to learn unified representations of interactions from mixed data. UniKT is a stack of Transformer decoder layers for handling long-term dependencies among students’ historical interactions and future performance. We evaluate UniKT on six publicly available real-world educational datasets, and experimental results demonstrate that our method outperforms the majority of existing DLKT models in terms of AUC and accuracy. Furthermore, the empirical analysis shows the strong transferability and adaptability of UniKT in learning from multiple sources. To encourage reproducible research, we make our data and code publicly available at https://pykt.org/ .

Bus passenger profiling is a critical task for optimizing urban transportation, but it is hindered by three key challenges: the heterogeneity of passenger behaviors, complex station-level interactions, and the prevalence of sparse, noisy transit data. Conventional end-to-end models that operate on aggregated traffic flow often fail to address these issues systematically. To overcome these limitations, this paper proposes GRASP, a novel two-stage paradigm for passenger profiling and flow prediction. In the first stage, GRASP acts as a disentangling module, constructing a passenger-centric graph to cluster individuals into distinct behavioral profiles based on their co-occurrence patterns. In the second stage, it performs profile-aware forecasting by learning group-specific, dynamic spatio-temporal dependencies using an adaptive station graph. This station-level model is further enhanced by a contrastive learning objective to ensure robustness against data imperfections. Extensive experiments on three real-world datasets demonstrate that GRASP not only achieves significantly superior flow prediction accuracy but also uncovers actionable passenger profiles. By structurally decoupling passenger behavior from station-level dynamics, GRASP offers a more interpretable and effective solution for data-driven public transportation management.

Knowledge tracing (KT) involves using the historical records of student-learning interactions to anticipate their performance on forthcoming questions. Central to this process is the modeling of human cognition to gain deeper insights into how knowledge is acquired and retained. Human cognition is characterized by two key features: long-term cognitive trends, reflecting the gradual accumulation and stabilization of knowledge over time, and short-term cognitive fluctuations, which arise from transient factors such as forgetting or momentary lapses in attention. Although existing attention-based KT models effectively capture long-term cognitive trends, they often fail to adequately address short-term cognitive fluctuations. These limitations lead to overly smoothed cognitive features and reduced model performance, especially when the test data length exceeds the training data length. To address these problems, we propose FlucKT, a novel short-term cognitive fluctuations enhanced attention network for KT tasks. FlucKT improves the attention mechanism in two ways: First, by using a decomposition-based layer with causal convolution to separate and dynamically reweight long-term and short-term cognitive features. Second, by introducing a kernelized bias attention score penalty to enhance focus on short-term fluctuations, improving length generalization capabilities. Our contributions are validated through extensive experiments on three real-world datasets, demonstrating significant improvements in length generalization and prediction performance.

Knowledge tracing (KT) is an essential task in online education systems. It aims to predict the future performance of students based on their historical learning interaction data. Despite significant advancements in attention-based KT models, they still face some limitations: inaccurate input representation and excessive student forgetting modeling. These limitations often lead to the attention noise problem: the model assigns non-negligible attention weight to some information that is cognitively irrelevant in nature, thereby generating interference signals. To address this problem, we propose a novel KT model, i.e., DenoiseKT. DenoiseKT effectively models the difficulty of the questions and utilizes graph neural network to capture the complex relationship between questions, thereby refining the representations of input features. Additionally, the denoised attention mechanism introduces a weight factor to reduce the model's attention weight distribution on irrelevant information. We extensively compare DenoiseKT with 22 state-of-the-art KT models on 4 widely-used public datasets. Experimental results show that DenoiseKT can effectively solve the attention noise problem and outperform other models. The source code of DenoiseKT is available at https://pykt.org.

Knowledge tracing (KT) aims to predict students' future performance based on their past learning activities. However, no one is perfect. Factors such as carelessness, fatigue, and stress often cause students to make mistakes on problems they have already mastered, leading to anomalies in their historical learning data. These anomalies disrupt inherent patterns in the data, misleading the KT model. Extracting cognitive patterns that accurately reflect students' knowledge mastery from such error-prone data remains a significant challenge. Against this background, this paper proposes a novel KT method named RoubstKT, inspired by educational measurement theory and frequency-based decomposition. A cognitive decoupling analyzer is proposed to decouple the student's cognitive pattern and random factors from the data through smoothing and subtraction operations, then recombine them using a gating mechanism or adaptive parameter fusion strategy. To more effectively diagnose students' knowledge mastery, we employ a decay-based attention mechanism that focuses on random behaviors at adjacent time steps. We conducted comprehensive experiments based on real-world datasets and targeted datasets with added random noise. The experimental results demonstrated the effectiveness of the proposed method.

Knowledge tracing (KT) involves utilizing historical data from students’ learning interactions to model their mastery of knowledge over time, with the aim of predicting their future performance in interactions. Recently, significant advancements have been achieved through the application of various deep learning methodologies to address the KT challenge. However, a considerable proportion of deep learning-based knowledge tracing (DLKT) approaches exhibit striking similarities in their methodologies, and model designs, and even the outcomes demonstrate minimal divergence. In addition, the evaluation procedures employed in current DLKT studies are not standardized, resulting in substantial inconsistencies in the reported area under the curve (AUC) outcomes, despite analyzing the same model on identical datasets. To address the two aforementioned problems, this paper proposes a generalized DLKT framework and represents the existing DLKT models with five components, i.e., multimodal data encoder, student knowledge memory, auxiliary knowledge base, learning outcome objective, and computational efficiency and scalability. Furthermore, we develop and open source a standardized DLKT benchmark platform namedpyKT,1that consists of a standardized set of integrated data preprocessing procedures on 9 popular datasets across different domains, and 21 frequently compared DLKT model implementations. WithpyKT, we conduct empirical and reproducible research to assess the performance of prevalent DLKT algorithms in an unbiased and clear setting over multiple data sources. Finally, we discuss the applications of KT techniques in the educational sector and their future development directions.

Attribute graph anomaly detection aims to identify nodes that significantly deviate from the majority of normal nodes, and has received increasing attention due to the ubiquity and complexity of graph-structured data in various real-world scenarios. However, current mainstream anomaly detection methods are primarily designed for centralized settings, which may pose privacy leakage risks in certain sensitive situations. Although federated graph learning offers a promising solution by enabling collaborative model training in distributed systems while preserving data privacy, a practical challenge arises as each client typically possesses a limited amount of graph data. Consequently, naively applying federated graph learning directly to anomaly detection tasks in distributed environments may lead to suboptimal performance results. We propose a federated graph anomaly detection framework via contrastive self-supervised learning (CSSL) [federated CSSL anomaly detection framework (FedCAD)] to address these challenges. FedCAD updates anomaly node information between clients via federated learning (FL) interactions. First, FedCAD uses pseudo-label discovery to determine the anomaly node of the client preliminarily. Second, FedCAD employs a local anomaly neighbor embedding aggregation strategy. This strategy enables the current client to aggregate the neighbor embeddings of anomaly nodes from other clients, thereby amplifying the distinction between anomaly nodes and their neighbor nodes. Doing so effectively sharpens the contrast between positive and negative instance pairs within contrastive learning, thus enhancing the efficacy and precision of anomaly detection through such a learning paradigm. Finally, the efficiency of FedCAD is demonstrated by experimental results on four real graph datasets.

Accurately judging student on-going performance is crucial for adaptive teaching. In this work, we focus on the task of automatically predicting students' levels of mastery of math questions from teacher-student classroom dialogue data in online one-on-one classes. As a step toward this direction, we introduce the Multi-turn Classroom Dialogue (MCD) dataset as a benchmark testing the capabilities of machine learning models in classroom conversation understanding of student performance judgment. Our dataset contains aligned multi-turn spoken language of 5000+ unique samples of solving grade-8 math questions collected from 500+ hours' worth of online one-on-one tutoring classes. In our experiments, we assess various state-of-the-art models on the MCD dataset, highlighting the importance of understanding multi-turn dialogues and handling noisy ASR transcriptions. Our findings demonstrate the dataset's utility in advancing research on automated student performance assessment. To encourage reproducible research, we make our data publicly available at https://github.com/ai4ed/MCD.

Knowledge tracing (KT) is a prediction task that aims to predict students’ future performance based on their past learning data. The rapid progress in attention mechanisms has led to the emergence of various high-performing attention based KT models. However, in online or personalized education settings, students’ varying learning paths result in different lengths of student interaction sequences, which poses a significant challenge for attention based KT models as their context window sizes are fixed during both training and prediction stages. We refer to this as the length extrapolation of KT model. In this paper, we propose extraKT to facilitate better extrapolation that learn from student interactions with a short context window and continue to perform well across various longer context window sizes at prediction stage. Specifically, we negatively bias attention scores with linearly decreasing penalties that are proportional to query-key distance, which efficiently represents short-term forgetting characteristics of student knowledge states. We conduct comprehensive and rigorous experiments on three real-world educational datasets. The results show that our extraKT model exhibits robust length extrapolation capability and outperforms state-of-the-art baseline models in terms of AUC and accuracy. To encourage reproducible research, we merge our data and code to the publicly available pyKT benchmark at https://github.com/pykt-team/pykt-toolkit.

The COVID-19 pandemic has posed great challenges to public health services, government agencies, and policymakers, raising huge social conflicts between public health and economic resilience. Policies such as reopening or closure of business activities are formulated based on scientific projections of infection risks obtained from infection dynamics models. Though most parameters in epidemic prediction service models can be set with domain knowledge of COVID-19, a key parameter, namely, human mobility, is often challenging to estimate due to complex spatio-temporal correlations and social contexts under escalating COVID-19 facilities. Moreover, how to integrate the various implicit features to accurately predict infectious cases is still an open issue. To address this challenge, we formulate the problem as a spatio-temporal network representation problem and propose STEP, a Spatio-Temporal Epidemic Prediction framework, to estimate pandemic infection risk of a city by integrating various real-world conditions (e.g., City Risk Index, climate, and medical conditions) into graph-structured data. We also employ a multi-head attention mechanism in representation learning to extract implicit features for a given city. Extensive experiments have been conducted upon the real-world dataset for 51 states (50 states and Washington, D.C.) of the USA. Experimental results show that STEP can yield more accurate pandemic infection risk estimation than baseline methods. Moreover, STEP outperforms other methods in both short-term and long-term prediction.

Region profiles are summaries of characteristics of urban regions. Region profiling is a process to discover the correlations between urban regions. The learned urban profiles can be used to represent and identify regions in supporting downstream tasks, e.g., region traffic status estimation. While some efforts have been made to model urban regions, representation learning with awareness of graph-structured data can improve the existing methods. To do this, we first construct an attribute spatio-temporal graph, in which a node represents a region, an edge represents mobility across regions, and a node attribute represents a region’s point of interest (PoI) distribution. The problem of region profiling is reformulated as a representation learning problem based on attribute spatio-temporal graphs. To solve this problem, we developed URGENT, a spatio-temporal graph learning framework. URGENT is made up of two modules. The graph convolutional neural network is used in the first module to learn spatial dependencies. The second module is an encoding–decoding temporal learning structure with self-attention mechanism. Furthermore, we use the learned representations of regions to estimate region traffic status. Experimental results demonstrate that URGENT outperforms major baselines in estimation accuracy under various settings and produces more meaningful results.

Mobile network operators (MNOs) allocate computing and caching resources for mobile users by deploying a central control system. Existing studies mainly use programming and heuristic methods to solve the resource allocation problem, which ignores the energy cost problem that is really significant to the MNO. To solve this problem, in this article, we design a joint computing and caching framework by integrating deep deterministic policy gradient (DDPG) algorithm. Especially, we focus on the Internet of Vehicles scenario, which needs the support of mobile network provided by MNO. We first formulate an optimization problem to minimize MNO’s energy cost by considering the computation and caching energy costs jointly. Then, we turn the formulated problem into a reinforcement learning problem and utilize DDPG methods to solve this problem. The final simulation result shows that our solution can reduce energy costs by more than 15%, while ensuring the tasks can be completed on time.

Traffic flow prediction is an integral part of an intelligent transportation system and thus fundamental for various traffic-related applications. Buses are an indispensable way of moving for urban residents with fixed routes and schedules, which leads to latent travel regularity. However, human mobility patterns, specifically the complex relationships between bus passengers, are deeply hidden in this fixed mobility mode. Although many models exist to predict traffic flow, human mobility patterns have not been well explored in this regard. To address this research gap and learn human mobility knowledge from this fixed travel behaviors, we propose a multi-pattern passenger flow prediction framework, MPGCN, based on Graph Convolutional Network (GCN). Firstly, we construct a novel sharing-stop network to model relationships between passengers based on bus record data. Then, we employ GCN to extract features from the graph by learning useful topology information and introduce a deep clustering method to recognize mobility patterns hidden in bus passengers. Furthermore, to fully utilize spatio-temporal information, we propose GCN2Flow to predict passenger flow based on various mobility patterns. To the best of our knowledge, this paper is the first work to adopt a multi-pattern approach to predict the bus passenger flow by taking advantage of graph learning. We design a case study for optimizing routes. Extensive experiments upon a real-world bus dataset demonstrate that MPGCN has potential efficacy in passenger flow prediction and route optimization.

License plate is an essential characteristic to identify vehicles for the traffic management, and thus, license plate recognition is important for Internet of Vehicles. Since 5G has been widely covered, mobile devices are utilized to assist the traffic management, which is a significant part of Industry 4.0. However, there have always been privacy risks due to centralized training of models. Also, the trained model cannot be directly deployed on the mobile device due to its large number of parameters. In this article, we propose a federated learning-based license plate recognition framework (FedLPR) to solve these problems. We design detection and recognition model to apply in the mobile device. In terms of user privacy, data in individuals is harnessed on their mobile devices instead of the server to train models based on federated learning. Extensive experiments demonstrate that FedLPR has high accuracy and acceptable communication cost while preserving user privacy.

Mobile crowdsourcing (MCS), as an effective and crucial technique of Industrial Internet of Things, is enabling smart city initiatives in the real world. It aims at incorporating the intelligence of dynamic crowds to collect and compute decentralized ubiquitous sensing data that can be used to solve major urbanization problems such as traffic congestion. The shared bus, as a neotype transportation mode, aims at improving the resource utilization rate and maintaining the advantages of convenience and economy. In this article, we provide a scheme to profile shared buses through heterogeneous mobile crowdsourced data (TRProfiling). First, we design an MCS-based shared bus data generation and collection solution to overcome the aforementioned data scarcity issue. Then, we propose a travel profiling to profile resident travel and design a method called multiconstraint evolution algorithm to optimize the routes. Experimental results demonstrate that TRProfiling has an excellent performance in satisfying passengers’ travel requirements.