Discrete latent representation techniques, such as Vector Quantization (VQ) and Sparse Coding (SC), have demonstrated superior image reconstruction and generation quality compared to continuous representation methods in Variational Autoencoders (VAEs). However, existing approaches often treat the latent representations of an image independently in their discrete representation space, neglecting both the inherent structural information within each representation and the correlations among them. This oversight leads to coarse representations and suboptimal generated results. In this paper, we address these limitations by introducing correlations among and within the latent representations of individual images in the latent discrete space of VAEs using sparse coding. We impose two-dimensional structural information through adaptive thresholding, enhancing local structure in image representations while suppressing noise via parsimonious representation with a learned dictionary. Empirical studies on three real benchmark datasets, including a clinical Ultrasound dataset, BSDS500, and mini-Imagenet, demonstrate that our proposed model preserves fine-grained details in image reconstruction and significantly outperforms baseline models of SC-VAE and VQ-VAE across objective and subjective image quality metrics. Particularly noteworthy are the substantial performance improvements observed on the ultrasound dataset, where structure information is crucial. Specifically, we observe significant performance improvements of 7.68 % and 17.03 % in SSIM, 3.25 dB and 6.58 dB in PSNR, 0.15 and 0.24 in LPIPS, 45.38 and 84.05 in FID over SC-VAE and VQ-VAE, respectively, indicating the superiority of our method in terms of image reconstruction quality and fidelity.

Current vision-language models may grasp basic spatial cues and simple directions (e.g. left, right, front, back), but struggle with the multi-dimensional spatial reasoning necessary for human-like understanding and real-world applications. To address this gap, we develop SPHERE (Spatial Perception and Hierarchical Evaluation of REasoning), a hierarchical evaluation framework supported by a new human-annotated dataset. SPHERE systematically probes models across increasing levels of complexity, from fundamental skills to multi-skill integration and high-level reasoning that combines spatial, visual, and logical understanding. Benchmark evaluation of state-of-the-art models reveals significant deficiencies, especially in reasoning about distance and proximity, understanding both egocentric and allocentric perspectives, and applying spatial logic in physical contexts. These findings expose critical blind spots in existing models and underscore the need for more advanced spatial reasoning techniques, driving the development of vision-language models that align more closely with human spatial cognition.

In this paper, the acoustic signal generated by the damaged water supply pipeline is deeply analyzed by Ansys simulation software. By constructing a three-dimensional model of water supply pipeline, the acoustic signals generated in the pipeline under different damage forms are simulated, and effective acoustic characteristic parameters are successfully extracted. In the research process, firstly, the three-dimensional model of water supply pipeline is established, and many common damage forms are simulated. Then, using the acoustic analysis module of Ansys software, the acoustic signal generated at the damaged part is simulated, and the simulation results are analyzed in detail by signal processing technology. By comparing the characteristic parameters of acoustic signals under different damage forms, the differences and laws between them are found, which provides a new theoretical basis for the damage detection of water supply pipelines.

Automating business collaboration promises efficiency and effectiveness increases with means of service-oriented cloud computing. For automating cross-organizationally connected business processes operation, existing choreography languages are typically technology driven and focus less on their socio-technical suitability and expressiveness that recognizes the interaction between people in organizations and technology in workplaces. The eSourcing Markup Language (eSML) addresses the gap, extracted from our studies of suitability and expressiveness for supporting the automation of business collaboration. eSML has been applied in case studies and its development approach is replicable for exploring strengths and weaknesses of other choreography languages.

The authors have previously shown dynamical properties-dynamics of the activities for states-for higher order random neural networks, which use the weighted sum of products of input variables, with the digital state {1-,1} model. The paper describes dynamical properties for higher order random neural networks with the analog state models and the digital state (0,1) model.