This study develops a novel theoretical framework for analyzing wave propagation in semiconductor media based on a generalized multi-temperature quantum photo-thermoelastic model. The formulation extends classical photo-thermoelastic semiconductor theory by incorporating nonequilibrium heat transport through a multi-temperature approach, in which conductive and thermodynamic temperatures interact via coupling parameters. Quantum carrier transport is introduced using a density-gradient model, leading to a higher-order diffusion term that captures nonlocal effects at micro- and nano-scales. The resulting system describes the fully coupled interaction among elastic deformation, multi-temperature heat conduction, and photogenerated carrier dynamics under laser excitation. A normal-mode analysis is employed to derive the governing dispersion relations and obtain analytical expressions for the physical fields. Numerical simulations, performed for silicon as a representative semiconductor, are used to examine the influence of multi-temperature parameters on temperature distribution, displacement, carrier density, and stress components. The results demonstrate that the multi-temperature formulation significantly modifies thermal wave behavior, enhances energy redistribution, and reduces stress concentration compared with classical single-temperature models. Moreover, the interplay between quantum diffusion and multi-temperature effects leads to smoother field profiles and increased attenuation. The proposed model provides new insights for the design of advanced optoelectronic and nano-scale semiconductor devices operating under ultrafast thermal loading.
Artificial Intelligence Generated Content (AIGC)-based artistic creation suffers from content distortion and excessive stylization due to weak multimodal fusion. This study proposes an Enhanced Cross-modal Attention Control (ECAC) model for controllable artistic image generation. Built on the Diffusion Transformer (DiT) backbone, the model integrates three-branch attention mechanisms to fuse text semantics, style embeddings, and content features. A learnable modality fusion weight matrix and temporal style intensity control strategy are introduced to balance style expression and content fidelity. Experiments on the ArtBench-10 dataset show that ECAC achieves an FID of 15.82, LPIPS of 0.247, and average style loss of 0.159, outperforming CycleGAN, Diffusion UNet, and Mamba-ST. Ablation studies validate the effectiveness of cross-modal attention and dynamic fusion. This study aims to provide a competitive pattern recognition scheme for AIGC art generation.
Traditional methods for egg fertility detection often suffer from low efficiency, high costs, and destructive sampling. While deep learning offers a nondestructive alternative, existing models struggle to balance robustness and efficiency, particularly when handling the subtle features of fertilized eggs under stringent deployment constraints. To address this, we propose a novel lightweight classification framework based on an improved YOLOv8 architecture. Our approach introduces three key innovations: a fully Ghost-based C2f (FG-C2f) module to eliminate backbone redundancy, an SCDown decoupled downsampling strategy to optimize deep-layer computation, and a minimalist depthwise convolution head for extreme parameter compression. Experimental results demonstrate that our proposed method significantly outperforms the baseline YOLOv8n-cls, achieving comparable or superior classification accuracy while reducing model parameters by over 60% and increasing inference speed by more than 40%. By achieving an optimal trade-off between precision and efficiency, this work provides a practical solution for embedded agricultural applications and offers valuable insights for designing efficient pattern recognition systems on resource-constrained edge devices.
Virtual communities are characterized by complex and dynamic interaction patterns among heterogeneous users and content. We focus on virtual communities because they present unique challenges, including diverse user types, multiple interaction modalities, and rapidly evolving network structures, which make predicting community-network evolution particularly complex and valuable. Accurately predicting the evolution of these interactions is crucial for effective community management and user engagement. In this paper, we propose DHGEP, a Dynamic Heterogeneous Graph Evolution Predictor, which leverages graph learning techniques to model and forecast the temporal evolution of virtual community interaction networks. DHGEP constructs a heterogeneous graph representing various user types, content, and interaction relationships, and employs a dynamic graph neural network enhanced with adaptive attention mechanisms to capture both structural and temporal dependencies. Experimental results on simulated and real-world community datasets demonstrate that DHGEP significantly outperforms baseline methods in predicting future interactions, providing a practical tool for proactive community management and insight-driven decision making.
