WLWeigang Lu

Weigang Lu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Weigang Lu has a total of 79 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Generative Agents for Learning Behavioral Simulation: An Interview-Driven and Multiexpert Reflection Framework

    by Cunling Bian, Zimeng Zhai, Saisai Ye, Weigang Lu

    Accurately modeling and predicting student learning behaviors remains a central challenge in educational research, as traditional approaches based on surveys or learning-process traces often struggle to capture the multidimensional and latent nature of learners' motivational and strategic states. Recent advances in Large Language Models (LLMs) and generative agents offer new opportunities to simulate learning behaviors by integrating memory construction, contextual reasoning, and expert-informed reflection. In this study, we propose a theory-grounded framework for Learning Behavioral Simulation that leverages structured interview transcripts and multi-expert reflective modeling to instantiate generative student agents. We conducted interviews with 46 preparatory mathematics students and generated domain-specific reflections, spanning educational psychology, learning sciences, educational technology, assessment, and sociocultural education, using LLM-instantiated educational expert agents. The instantiated generative student agents, informed by these expert reflections, were tasked with predicting responses on the Motivated Strategies for Learning Questionnaire, enabling construct-level evaluation of motivational and learning strategy behaviors. Experimental results show that interview-driven agents augmented with expert reflections consistently outperform persona- and demographic-based baselines. Further analysis reveals that reflections from different expert domains capture complementary dimensions of learning behavior, improving both predictive robustness and interpretability. Overall, this work demonstrates how interview-grounded memory construction and expert-guided reflection can support interpretable, scalable, and model-agnostic generative simulation of student learning behaviors, with implications for personalized and explainable educational applications.

  • Multimodal Classroom Climate Recognition: Dataset, Method, Application

    by Weigang Lu, Ying Chen, Yang Yang, Runfei Song, Tao Wang, Cunling Bian

    Classroom climate is a socio-emotional environment shaped by teacher–student interactions, directly influencing classroom practices and student development. However, the diverse behaviors, subtle emotional cues, and complex interaction structures present in real classrooms make it difficult to develop intelligent models that can accurately perceive and interpret this environment. Existing methods often rely on single-modality features or generic multimodal models, and thus fail to represent the distinctive characteristics of classroom groups, resulting in shallow or unstable climate recognition. To effectively capture the dynamic, multimodal patterns of real-world classrooms, we introduce the Multimodal Classroom Climate Dataset (MCCD), which includes a wide range of instructional activities, and propose the Multimodal Classroom Climate Recognition Network (MCCRNet). The network integrates visual cues such as skeletal actions with auditory features such as emotional tone to construct cross-modal representations that align with the logic of classroom interactions. Experimental results indicate that the proposed approach exceeds existing methods, laying a preliminary groundwork for advancing the understanding of classroom ecology and instructional decision-making.


  • LightDTA: lightweight drug-target affinity prediction via random-walk network embedding and knowledge distillation

    by Xiaoyu Huang, Xiangpeng Bi, Nianwen Xing, Wenjian Ma, Huasen Jiang, Qing Cai, Weigang Lu, Fei Yang, Zhiqiang Wei, Shugang Zhang


  • Open Access

    Rehearsal-Free Continual Learning for Emerging Unsafe Behavior Recognition in Construction Industry

    by Tao Wang, Saisai Ye, Zimeng Zhai, Weigang Lu, Cunling Bian

    In the realm of Industry 5.0, the incorporation of Artificial Intelligence (AI) in overseeing workers, machinery, and industrial systems is essential for fostering a human-centric, sustainable, and resilient industry. Despite technological advancements, the construction industry remains largely labor intensive, with site management and interventions predominantly reliant on manual judgments, leading to inefficiencies and various challenges. This research emphasizes identifying unsafe behaviors and risks within construction environments by employing AI. Given the continuous emergence of unsafe behaviors that requires certain caution, it is imperative to adapt to these novel categories while retaining the knowledge of existing ones. Although deep convolutional neural networks have shown excellent performance in behavior recognition, they traditionally function as predefined multi-way classifiers, which exhibit limited flexibility in accommodating emerging unsafe behavior classes. Addressing this issue, this study proposes a versatile and efficient recognition model capable of expanding the range of unsafe behaviors while maintaining the recognition of both new and existing categories. Adhering to the continual learning paradigm, this method integrates two types of complementary prompts into the pre-trained model: task-invariant prompts that encode knowledge shared across tasks, and task-specific prompts that adapt the model to individual tasks. These prompts are injected into specific layers of the frozen backbone to guide learning without requiring a rehearsal buffer, enabling effective recognition of both new and previously learned unsafe behaviors. Additionally, this paper introduces a benchmark dataset, Split-UBR, specifically constructed for continual unsafe behavior recognition on construction sites. To rigorously evaluate the proposed model, we conducted comparative experiments using average accuracy and forgetting as metrics, and benchmarked against state-of-the-art continual learning baselines. Results on the Split-UBR dataset demonstrate that our method achieves superior performance in terms of both accuracy and reduced forgetting across all tasks, highlighting its effectiveness in dynamic industrial environments.

  • Top Cited Papers
    Open Access

    Effects of AI-generated images in visual art education on students' classroom engagement, self-efficacy and cognitive load

    by Cunling Bian, Xiaofang Wang, Yingxue Huang, Shan Zhou, Weigang Lu


  • SSPPI: Cross-Modality Enhanced Protein–Protein Interaction Prediction From Sequence and Structure Perspectives

    by Xiangpeng Bi, Wenjian Ma, Huasen Jiang, Weigang Lu, Zhiqiang Wei, Shugang Zhang

    Recent advances have shown great promise in mining multimodal protein knowledge for better protein–protein interaction (PPI) prediction by enriching the representation of proteins. However, existing solutions lack a comprehensive consideration of both local patterns and global dependencies in proteins, hindering the full exploitation of modal information. Additionally, the inherent disparities between modalities are often disregarded, which may lead to inferior modality complementarity effects. To address these issues, we propose a cross-modality enhanced PPI prediction method from the perspectives of protein sequence and structure modalities, namely SSPPI. In this framework, our main contribution is that we integrate both sequence and structural modalities of proteins and employ an alignment and fusion method between modalities to further generate more comprehensive protein representations for PPI prediction. Specifically, we design two modal representation modules (Convformer and Graphormer) tailored for protein sequence and structure modalities, respectively, to enhance the quality of modal representation. Subsequently, we introduce a Cross-modality enhancer module to achieve alignment and fusion between modalities, thereby generating more informative modal joint representations. Finally, we devise a cross-protein fusion (CPF) module to model residue interaction processes between proteins, thereby enriching the joint representation of protein pairs. Extensive experimentation on four benchmark datasets demonstrates that our proposed model surpasses all current state-of-the-art (SOTA) methods. The source codes are publicly available at the following link https://github.com/bixiangpeng/SSPPI/


  • Geometric Deep Learning for Protein-Ligand Affinity Prediction With Hybrid Message Passing Strategies

    by Jiaren Li, Huasen Jiang, Wenjian Ma, Xiangpeng Bi, Rui Chen, Weigang Lu, Qing Cai, Fei Yang, Zhiqiang Wei, Shugang Zhang

    Accurate prediction of protein-ligand affinity (PLA) is critical for drug discovery. Recent deep learning approaches have adopted data-driven models for PLA prediction by learning intrinsic patterns from one-dimensional (1D) sequential or two-dimensional (2D) graph representations of proteins and ligands. However, these low-dimensional methods overlook the three-dimensional (3D) geometric features, which are hypothesized to be critical in binding interaction. To address the above problem, we present a Geometric deep learning approach with Hybrid message passing strategies-HybridGeo, for protein-ligand affinity prediction. We adopt dual-view graph learning to model the intra- and inter-molecular atomic interactions and propose to aggregate the spatial information with hybrid strategies. In addition, to fully model the inter-residue dependency upon message aggregation, we adopt a geometric graph transformer on the residue-scale graph of protein pockets. Extensive experiments on the PDBbind dataset show that HybridGeo achieves state-of-the-art performance with a Root Mean Square Error (RMSE) of 1.172. HybridGeo also achieves the best among all baseline models on three external test sets, showcasing good generalizability and robustness. Through systematic ablation experiments, we validated the effectiveness of the proposed modules, and further demonstrated the superior performance of HybridGeo in predicting the binding affinity of macrocyclic compound complexes through case studies. Visualization analysis further indicates the biological interpretability of the model predictions. Our code is publicly available at https://github.com/anxiangbiye1231/HybridGeo


  • Open Access

    Spatial–Temporal Heatmap Masked Autoencoder for Skeleton-Based Action Recognition

    by Cunling Bian, Yang Yang, Tao Wang, Weigang Lu

    Skeleton representation learning offers substantial advantages for action recognition by encoding intricate motion details and spatial–temporal dependencies among joints. However, fully supervised approaches necessitate large amounts of annotated data, which are often labor-intensive and costly to acquire. In this work, we propose the Spatial–Temporal Heatmap Masked Autoencoder (STH-MAE), a novel self-supervised framework tailored for skeleton-based action recognition. Unlike coordinate-based methods, STH-MAE adopts heatmap volumes as its primary representation, mitigating noise inherent in pose estimation while capitalizing on advances in Vision Transformers. The framework constructs a spatial–temporal heatmap (STH) by aggregating 2D joint heatmaps across both spatial and temporal axes. This STH is partitioned into non-overlapping patches to facilitate local feature learning, with a masking strategy applied to randomly conceal portions of the input. During pre-training, a Vision Transformer-based autoencoder equipped with a lightweight prediction head reconstructs the masked regions, fostering the extraction of robust and transferable skeletal representations. Comprehensive experiments on the NTU RGB+D 60 and NTU RGB+D 120 benchmarks demonstrate the superiority of STH-MAE, achieving state-of-the-art performance under multiple evaluation protocols.