Network Diffusion Research Articles

Precision-machine learning for the matrix element method

The matrix element method is the LHC inference method of choice for limited statistics. We present a dedicated machine learning framework, based on efficient phase-space integration, a learned acceptance and transfer function. It is based on a choice of INN and diffusion networks, and a transformer to solve jet combinatorics. We showcase this setup for the CP-phase of the top Yukawa coupling in associated Higgs and single-top production.

Diffusion network with spatial channel attention infusion and frequency spatial attention for brain tumor segmentation.

Accurate segmentation of gliomas is crucial for diagnosis, treatment planning, and evaluating therapeutic efficacy. Physicians typically analyze and delineate tumor regions in brain magnetic resonance imaging (MRI) images based on personal experience, which is often time-consuming and subject to individual interpretation. Despite advancements in deep learning technology for image segmentation, current techniques still face challenges in clearly defining tumor boundary contours and enhancing segmentationaccuracy. To address these issues, this paper proposes a conditional diffusion network (SF-Diff) with a spatial channel attention infusion (SCAI) module and a frequency spatial attention (FSA) mechanism to achieve accurate segmentation of the whole tumor (WT) region in braintumors. SF-Diff initially extracts multiscale information from multimodal MRI images and subsequently employs a diffusion model to restore boundaries and details, thereby enabling accurate brain tumor segmentation (BraTS). Specifically, a SCAI module is developed to capture multiscale information within and between encoder layers. A dual-channel upsampling block (DUB) is designed to assist in detail recovery during upsampling. A FSA mechanism is introduced to better match the conditional features with the diffusion probability distribution information. Furthermore, a cross-model loss function has been implemented to supervise the feature extraction of the conditional model and the noise distribution of the diffusionmodel. The dataset used in this paper is publicly available and includes 369 patient cases from the Multimodal BraTS Challenge 2020 (BraTS2020). The conducted experiments on BraTS2020 demonstrate that SF-Diff performs better than other state-of-the-art models. The method achieved a Dice score of 91.87%, a Hausdorff 95 of 5.47 mm, an IoU of 84.96%, a sensitivity of 92.29%, and a specificity of 99.95% onBraTS2020. The proposed SF-Diff performs well in identifying the WT region of the brain tumors compared to other state-of-the-art models, especially in terms of boundary contours and non-contiguous lesion regions, which is clinically significant. In the future, we will further develop this method for brain tumor three-class segmentation task.

A Survey of Deep Learning-Based Information Cascade Prediction

Online social media have significantly boosted the creation and transmission of information, accelerating the dissemination and interaction of vast amounts of data, thereby making the prediction of information cascades increasingly important. In recent years, deep learning has been extensively applied in the domain of information cascade prediction. This paper primarily classifies, organizes, and summarizes the current research status and classic algorithms of information cascade prediction methods based on deep learning. According to the different focuses on characterizing information cascade features, studies on deep learning-based information cascade prediction are classified from two perspectives, i.e., prediction targets and prediction methods. Each category is explained in detail, along with its principles, advantages, and disadvantages, and the commonly used datasets and evaluation metrics in this field are introduced. Additionally, this paper explores the role of symmetry in the structural patterns of information diffusion networks, analyzing how symmetry impacts the pathways and efficiency of information dissemination. Finally, this paper summarizes the potential future research directions and development trends in this domain.

The research of knowledge diffusion network model for Tourism Destination-Public ecological civilization.

Visitor education plays a crucial role in the knowledge diffusion process in outdoor recreation and nature-based tourism. It entails sharing information, experiences, and insights with visitors to enhance their understanding and appreciation of the natural environment. Our methodology for investigating the diffusion of ecological civilization knowledge in tourism destinations involves constructing a knowledge diffusion network model. In this model, scenic spots, tourists, and the public are defined as network nodes, with the communication channels between them representing the edges of the network. By constructing a scale-free complex network, the knowledge diffusion mode of scenic spots can be depicted. The layer of resource supply node consists of different scenic spots, forming the core nodes set for knowledge diffusion in the tourism industry. This research aims to further explore the social and economic value of the tourism areas' ecological civilization knowledge diffusion, as well as analyze the path, quantity, and cost of knowledge diffusion. by analyzing this knowledge diffusion network model, insights into the effectiveness and impact of visitor education in promoting ecological civilization and sustainable practices in tourism destinations can be gained. Overall, this approach provides a theoretical framework for investigating and comprehending the knowledge diffusion process in Tourism Destination-Public ecological civilization, thereby shedding light on the social and economic benefits that can be derived from sustainable tourism practices.

Option Pricing Based on Neural Stochastic Differential Equations

Firstly, based on the Black-Scholes stock price model, the neural stochastic differential equation (NSDE) model was established by parameterizing the asset return rate and volatility as a drift network and a diffusion network, respectively. Secondly, in the empirical analysis, the underlying asset as a single stock option was used as the research object, and real stock data was used for the network training and testing. The experimental results show that the NSDE model can overcome the defects of the constant assumption of the Black-Scholes model. Finally, for the case where the price of the underlying asset of the option was unobservable, we proposed that the price of any target option and the price of a known option could be constrained within the Wasserstein distance of their risk-neutral equivalent martingale measure, and theoretically proved the method.

De novo protein design with a denoising diffusion network independent of pretrained structure prediction models.

The recent success of RFdiffusion, a method for protein structure design with a denoising diffusion probabilistic model, has relied on fine-tuning the RoseTTAFold structure prediction network for protein backbone denoising. Here, we introduce SCUBA-diffusion (SCUBA-D), a protein backbone denoising diffusion probabilistic model freshly trained by considering co-diffusion of sequence representation to enhance model regularization and adversarial losses to minimize data-out-of-distribution errors. While matching the performance of the pretrained RoseTTAFold-based RFdiffusion in generating experimentally realizable protein structures, SCUBA-D readily generates protein structures with not-yet-observed overall folds that are different from those predictable with RoseTTAFold. The accuracy of SCUBA-D was confirmed by the X-ray structures of 16 designed proteins and a protein complex, and by experiments validating designed heme-binding proteins and Ras-binding proteins. Our work shows that deep generative models of images or texts can be fruitfully extended to complex physical objects like protein structures by addressing outstanding issues such as the data-out-of-distribution errors.

TFDNet: A triple focus diffusion network for object detection in urban congestion with accurate multi-scale feature fusion and real-time capability

Vehicle detection in congested urban scenes is essential for traffic control and safety management. However, the dense arrangement and occlusion of multi-scale vehicles in such environments present considerable challenges for detection systems. To tackle these challenges, this paper introduces a novel object detection method, dubbed the triple focus diffusion network (TFDNet). Firstly, the gradient convolution is introduced to construct the C2f-EIRM module, replacing the original C2f module, thereby enhancing the network’s capacity to extract edge information. Secondly, by leveraging the concept of the Asymptotic Feature Pyramid Network on the foundation of the Path Aggregation Network, the triple focus diffusion module structure is proposed to improve the network’s ability to fuse multi-scale features. Finally, the SPPF-ELA module employs an Efficient Local Attention mechanism to integrate multi-scale information, thereby significantly reducing the impact of background noise on detection accuracy. Experiments on the VisDrone 2021 dataset reveal that the average detection accuracy of the TFDNet algorithm reached 38.4%, which represents a 6.5% improvement over the original algorithm; similarly, its mAP50:90 performance has increased by 3.7%. Furthermore, on the UAVDT dataset, the TFDNet achieved a 3.3% enhancement in performance compared to the original algorithm. TFDNet, with a processing speed of 55.4 FPS, satisfies the real-time requirements for vehicle detection.

A charge-preserving method for solving graph neural diffusion networks

The aim of this paper is to give a systematic mathematical interpretation of the diffusion problem on which Graph Neural Networks (GNNs) models are based. The starting point of our approach is a dissipative functional leading to dynamical equations which allows us to study the symmetries of the model. We provide a short review of graph theory and its relation with network σ-models adapted to our analysis. We discuss the conserved charges and provide a charge-preserving numerical method for solving the dynamical equations. In any dynamical system and also in GRAph Neural Diffusion (GRAND), knowing the charge values and their conservation along the evolution flow could provide a way to understand how GNNs and other networks work with their learning capabilities.

The international dissemination of China’s English academic journals on Twitter: Structure, theme and impact

Objective/Significance: To measure the international communication capacity of China’s English academic journal papers based on social media platforms. Method/Process: Social network analysis, content analysis and emotion analysis were used to describe the diffusion structure of China’s English academic journal papers on Twitter from the aspects of network structure and identify important node users, explore the identity characteristics and emotion characteristics of users as well as diffusion capability of journals of different disciplines. Conclusion/Discovery: Journal articles in the fields of natural sciences and engineering as well as humanities and social sciences exhibit a strong positive correlation between their diffusion structure and network size. Networks with a mixed diffusion structure typically have larger network sizes. The users occupying key positions in the diffusion network typically have profile information that is academically relevant. The retweets and comments are mainly statements without emotional inclination, and the comments are mainly summary descriptions of the papers. The distribution of diffusion width, diffusion strength and diffusion speed of each discipline is significantly different. There is a significant positive correlation between the diffusion structure of papers and their diffusion capabilities. Innovation/Value: To reveal the diffusion structure and characteristics of China’s English academic journal papers on social media platforms, so as to provide reference for expanding the visibility of China’s English academic journals and improving the international dissemination capacity of China’s English journals.

Does Parkinson's disease affect verb production in picture descriptions?

Research on Parkinson's disease (PD) has documented significant deficits in verb production, with more robust results in single word retrieval tasks than in connected speech, yet the underlying causes of these deficits are disputable, especially concerning connected speech production. We analyzed picture descriptions provided by 48 individuals with PD and 48 age-matched healthy controls, and examined the percent of nouns and verbs of all words, the number of described events, verbs denoting activity, verbs in active morpho-syntactic patterns, and transitive verbs. Individuals with PD produced a lower percent of verbs than did control participants, but the groups differed in no other variable. Scores on a cognitive screening task associated with the percent of verbs and the number of events. We suggest that verb retrieval in connected speech in PD reflects no specific difficulty with action semantics, but rather the spread of PD pathology into more diffuse verb-specific neural networks.

Studies on highly stable Na7Fe3(P2O7)4 as anode materials for lithium-ion batteries and migration behavior of lithium ions

In this paper, Na7Fe3(P2O7)4 target material is firstly evaluated as potential anode material for lithium ion batteries. It is found to have good electro-chemical performance and high stability as anode material for lithium batteries. It shows a capacity of 290 mAh g−1 for 200 cycles and increases to 388 mAh g−1 for 1000 cycles at a current density of 150 mA g−1. The Coulomb efficiency is maintained at about 98 % during the long cycling. The ex-situ experimental results indicate that Na7Fe3(P2O7)4 has high structural stability and the changes of cell volume and cell parameters are <0.62 % during charging and discharging. The results of GITT data show that the diffusion coefficients range from 3.15 × 10−11 cm2 S−1 to 1.12 × 10−12 cm2 S−1.The Li+/Na+ diffusion paths are estimated by the Bond valence-energy landscape (BEL) method and the exact energy barriers and minimal energy paths (MEPs) are determined using the nudged elastic band (NEB) method implemented in density functional theories (DFT). A two dimensional Na+ diffusion network is observed and Na+ diffusion along c-axis is prohibited. For the initial discharge stage, Na+ migrates in the Na+ transportation channel with an energy barrier of 0.48 eV along a-axis and 0.46 eV along b-axis. For the discharge later on, the intercalated Li+ enters the nearby Na+ vacancies and also migrate in the Na+ channel except Li+ at 8f Lid sites, which will stay where it is firstly introduced because of the higher energy barrier of 0.75 eV. The immobility of Lid is one of the important reasons for the irreversible capacity loss after the initial charge/discharge cycle.

Mining the Sun: Coal-to-solar transitions and energetic place-making in Appalachia

A striking effort is under way to use energy investment, a key feature of US federal climate policy, to build solar farms on former mining lands and in legacy mining communities in central Appalachia. Focusing on community solar initiatives, we consider the geosociality of energy transitions from the perspective of communities where many people have considerable knowledge of energy systems and how they work. We note that the politics of place-making in the context of environmental sacrifice and capital abandonment enables negotiations around energy that often seek to bridge deep social contradictions. We highlight how solar energy’s diffuse networks of power, both social and electrical, privilege forms of temporal labor that are alliance-building, compositionist, para-ethnographic and intimately concerned with the capacity to build relations. Vernacular energy expertise can play a major role in creating novel forms of political sociality. In West Virginia, this both builds on and significantly shifts the emphasis of long legacies of energetic place-making. We expect the ongoing development of community-led solar investment to contribute to substantially important diversifying effects in West Virginian communities, especially when these efforts can more explicitly address the state’s extreme geographical, race, and class-based inequalities.

An Optimized Object Detection Algorithm for Marine Remote Sensing Images

In order to address the challenge of the small-scale, small-target, and complex scenes often encountered in offshore remote sensing image datasets, this paper employs an interpolation method to achieve super-resolution-assisted target detection. This approach aligns with the logic of popular GANs and generative diffusion networks in terms of super-resolution but is more lightweight. Additionally, the image count is expanded fivefold by supplementing the dataset with DOTA and data augmentation techniques. Framework-wise, based on the Faster R-CNN model, the combination of a residual backbone network and pyramid balancing structure enables our model to adapt to the characteristics of small-target scenarios. Moreover, the attention mechanism, random anchor re-selection strategy, and the strategy of replacing quantization operations with bilinear interpolation further enhance the model’s detection capability at a low cost. Ablation experiments and comparative experiments show that, with a simple backbone, the algorithm in this paper achieves a mAP of 71.2% on the dataset, an improvement in accuracy of about 10% compared to the Faster R-CNN algorithm.

Ternary intermediate phase mediated high energy density and kinetically accelerated Sb-Cd electrode for liquid metal batteries

Liquid metal battery (LMB) is an attractive technology that can address the big challenge of efficient utilization of renewable energies, ascribed to its potential ultralong service lifetime, ideal cost-efficiency, and high safety. However, breaking through the energy density bottleneck remains a persistent challenge, due to the limited lithiation voltage and specific capacity of cathodes. Herein, we design a novel dual-active Sb-Cd cathode, which shows a new ternary alloying discharge mechanism with reversible formation of ternary intermediate phase LiCdSb, accompanied by a high discharge voltage of 1.0 V. Coupled with the excellent Li storage capability of Cd, the Sb80Cd20 cathode attains an attractive capacity (> 500 mAh g-1), outperforming all reported LMB cathodes. More importantly, upon deep lithiation, the conversion reaction of ternary LiCdSb to Li3Sb mediates the regeneration and dispersion of Cd melt in the cathode, which constructs in situ fast electron and lithium diffusion networks for the subsequent discharge, significantly accelerating the electrode reaction kinetics. Thus, the Sb80Cd20 electrode achieves an exceptional combination of high energy density (398.4 Wh kg-1 at 200 mA cm-2) and superior rate capability (542.5 W kg-1 at 2400 mA cm-2). Additionally, the Li||Sb-Cd battery exhibits an outstanding tolerance and self-healing ability towards thermal runaway.

LFDT-Fusion: A latent feature-guided diffusion Transformer model for general image fusion

For image fusion tasks, it is inefficient for the diffusion model to iterate multiple times on the original resolution image for feature mapping. To address this issue, this paper proposes an efficient latent feature-guided diffusion model for general image fusion. The model consists of a pixel space autoencoder and a compact Transformer-based diffusion network. Specifically, the pixel space autoencoder is a novel UNet-based latent diffusion strategy that compresses inputs into a low-resolution latent space through downsampling. Simultaneously, skip connections transfer multiscale intermediate features from the encoder to the decoder for decoding, preserving the high-resolution information of the original input. Compared to the existing VAE-GAN-based latent diffusion strategy, the proposed UNet-based strategy is significantly more stable and generates highly detailed images without adversarial optimization. The Transformer-based diffusion network consists of a denoising network and a fusion head. The former captures long-range diffusion dependencies and learns hierarchical diffusion representations, while the latter facilitates diffusion feature interactions to comprehend complex cross-domain information. Moreover, improvements to the diffusion model in noise level, denoising steps, and sampler selection have yielded superior fusion performance across six image fusion tasks. The proposed method illustrates qualitative and quantitative advantages, as evidenced by experimental results in both public datasets and industrial environments. The code is available at: https://github.com/BOYang-pro/LFDT-Fusion.

Differentiating broadcast from viral: a causal inference approach for information diffusion analysis

Classifying information diffusion patterns is critical to many information analysis areas, e.g., misleading information detection. However, diffusion pattern classification remains challenging when multiple users are involved. To address this challenge, this study aims to classify how information diffuses, distinguishing between broadcast and viral spreading, solely through the analysis of observational data from retweet networks on X (formerly known as Twitter). In broadcasting, most users directly receive information. However, viral spreading allows users the opportunity to receive information from a variety of sources. Therefore, viral spreading increases the likelihood of identifying misleading information. Existing methods classify diffusion types mainly through structural virality, which relies on the average distance between the users. However, when dealing with diffusion networks involving two or more information sources, these approaches can potentially lead to confusion regarding causality. To tackle this problem, we develop a deterministic causal inference method for categorizing information diffusion types. To the best of our knowledge, this is the first study investigating information diffusion types based on causality. This approach can be used to assess source credibility and assist in detecting misleading information. It can also be extended to other social media platforms.Graphical

VCC-DiffNet: Visual Conditional Control Diffusion Network for Remote Sensing Image Captioning

Pioneering remote sensing image captioning (RSIC) works use autoregressive decoding for fluent and coherent sentences but suffer from high latency and high computation costs. In contrast, non-autoregressive approaches improve inference speed by predicting multiple tokens simultaneously, though at the cost of performance due to a lack of sequential dependencies. Recently, diffusion model-based non-autoregressive decoding has shown promise in natural image captioning with iterative refinement, but its effectiveness is limited by the intrinsic characteristics of remote sensing images, which complicate robust input construction and affect the description accuracy. To overcome these challenges, we propose an innovative diffusion model for RSIC, named the Visual Conditional Control Diffusion Network (VCC-DiffNet). Specifically, we propose a Refined Multi-scale Feature Extraction (RMFE) module to extract the discernible visual context features of RSIs as input of the diffusion model-based non-autoregressive decoder to conditionally control a multi-step denoising process. Furthermore, we propose an Interactive Enhanced Decoder (IE-Decoder) utilizing dual image–description interactions to generate descriptions finely aligned with the image content. Experiments conducted on four representative RSIC datasets demonstrate that our non-autoregressive VCC-DiffNet performs comparably to, or even better than, popular autoregressive baselines in classic metrics, achieving around an 8.22× speedup in Sydney-Captions, an 11.61× speedup in UCM-Captions, a 15.20× speedup in RSICD, and an 8.13× speedup in NWPU-Captions.

An image information fusion based simple diffusion network leveraging the segment anything model for guided attention on thermal images producing colorized pedestrian masks

Because of the synergistic characteristics of the data they provide, thermal imagers are becoming more and more common as secondary data-collecting modules to complement classical optical imagers. Their application is especially beneficial when ambient elements like light, smoke, or other particulates need to be handled. As a result, fusion-based detection technology, like assistive driving systems, is trying to incorporate thermal imagers. In this work, we demonstrate that it is possible to create a mask for pedestrians in thermal imaging by building a basic Denoising Diffusion Probabilistic Model and using the results from the Segment Anything Model. Moreover, we show how the anticipated mask can fuse RGB and thermal pictures into a 3-channel output. Finally, we demonstrate how the loss function optimizes the end-to-end analysis in this field. Results from both the quantitative and qualitative analyses validate our methodology.

StarCAN-PFD: An Efficient and Simplified Multi-Scale Feature Detection Network for Small Objects in Complex Scenarios

Small object detection in traffic sign applications often faces challenges like complex backgrounds, blurry samples, and multi-scale variations. Existing solutions tend to complicate the algorithms. In this study, we designed an efficient and simple algorithm network called StarCAN-PFD, based on the single-stage YOLOv8 framework, to accurately recognize small objects in complex scenarios. We proposed the StarCAN feature extraction network, which was enhanced with the Context Anchor Attention (CAA). We designed the Pyramid Focus and Diffusion Network (PFDNet) to address multi-scale information loss and developed the Detail-Enhanced Conv Shared Detect (DESDetect) module to improve the recognition of complex samples while keeping the network lightweight. Experiments on the CCTSDB dataset validated the effectiveness of each module. Compared to YOLOv8, our algorithm improved mAP@0.5 by 4%, reduced the model size to less than half, and demonstrated better performance on different traffic sign datasets. It excels at detecting small traffic sign targets in complex scenes, including challenging samples such as blurry, low-light night, occluded, and overexposed conditions, showcasing strong generalization ability.

Realistic Facial Age Transformation with 3D Uplifting

AbstractWhile current facial re‐ageing methods can produce realistic results, they purely focus on the 2D age transformation. In this work, we present an approach to transform the age of a person in both facial appearance and shape across different ages while preserving their identity. We employ an α‐(de)blending diffusion network with an age‐to‐α transformation to generate coarse structure changes, such as wrinkles. Additionally, we edit biophysical skin properties, including melanin and hemoglobin, to simulate skin color changes, producing realistic re‐ageing results from ages 10 to 80 years. We also propose a geometric neural network that alters the coarse scale facial geometry according to age, followed by a lightweight and efficient network that adds appropriate skin displacement on top of the coarse geometry. Both qualitative and quantitative comparisons show that our method outperforms current state‐of‐the‐art approaches.