Explore the vast range of titles available.

Graph anomaly detection (GAD) is pivotal for security-critical applications like cybersecurity and financial fraud detection. While reconstruction-based Graph Neural Networks (GNNs) are prevalent, their efficacy is often compromised by two phenomena: (1) anomaly overfitting, where expressive models capture anomalous patterns, and (2) homophily-induced attenuation, where message passing smooths localized anomaly cues. This paper proposes CLEAN-GAD, a contamination-aware framework that mitigates anomaly influence during training through multi-view robust learning. Specifically, we develop a contrastive augmentation module that utilizes local inconsistency scores to identify and suppress pseudo-anomalous nodes and edges, thereby yielding a purified augmented view. To capture diverse anomaly signals, a frequency-adaptive encoder with dual-pass channels is designed to integrate low- and high-frequency information. Furthermore, we introduce a distribution-separation regularizer and cross-view alignment to stabilize learning and resolve view shifts. Theoretical analysis confirms that reducing the contamination ratio ρ expands the reconstruction-risk gap between normal and anomalous nodes, inherently boosting detection performance. Extensive experiments on multiple benchmark datasets from various domains demonstrate the superior anomaly detection performance of CLEAN-GAD.

Anonymous communication networks such as the Invisible Internet Project (I2P) are essential for safeguarding privacy and ensuring freedom of expression, necessitating robust performance and security evaluation in controlled environments. Network testbeds offer a reliable alternative to real-world testing. This paper proposes a dynamic modeling framework based on Graph Attention Network (GAT). We introduce a Region-Centric Initialization (RCI) strategy to establish an initial observation anchor, followed by a GAT-based Locality-Aware (GAT-LA) sampling mechanism that treats representative node selection as a dynamic learning task. Experimental results demonstrate that the GAT-LA mechanism significantly outperforms static methods in maintaining long-term similarity to real-world I2P performance metrics. The integrated stability penalty mechanism effectively suppresses excessive topological fluctuations, ensuring temporal smoothness across evolutionary cycles. Furthermore, the RCI strategy provides high engineering flexibility by supporting both automated scoring and target-oriented manual configuration. This paper presents a scalable methodology for dynamic network simulation with enhanced statistical alignment, providing a practical reference for security research within resource-constrained anonymous network ranges or testbeds.

Background Chrysosplenetin (CHR), a polymethoxy flavonol co-occurring with artemisinin (ART) in Artemisia annua L., reverses ART resistance in Plasmodium berghei K173 potentially by downregulating intestinal P-glycoprotein (P-gp, encoded by  Mdr1a ) expression. In the present study, we further elaborated on the mechanism by comparing differences in antimalarial activity and resistance-associated molecular expression profiles between ART alone and combination therapy in blood and tissues of Mdr1a wild-type (WT) and knockout (KO) mice infected with either sensitive or resistant malarial parasites. Methods We evaluated the effects of monotherapy and combination therapy in WT and KO mice infected with sensitive and resistant P. berghei K173 strains. The mRNA expressions of multi-resistance proteins (Mrp1, 2, 4, 5) and breast cancer resistance proteins (Bcrp) were detected. Hemoglobin levels, mRNA expressions of cytokines including tumor necrosis factor-α (IFN-α), interferon-α (IFN-α), and interleukin (IL-1β) in blood and tissues, and redox balance (ROS/GSH levels), as well as gene or protein expression of signaling pathway (PI3K/AKT-mTOR and MAPK) were investigated. Results In drug-resistant mice, combination therapy maintained the highest survival (100%) and inhibition (30%) rates and the lowest parasitaemia percentage (approximately 20.0%), irrespective of Mdr1a gene status. Furthermore, combination reshaped the spatial and ART resistance-phenotypic disparities in Mrps and Bcrp mRNA expressions (with a fold change ranging from 1.35 to 38.03), ROS/GSH balance (ranging from 1.02-fold to 10.18-fold), hemoglobin levels (ranging from 1.04-fold to 1.20-fold), and cytokine profiles (ranging from 1.14-fold to 37.79-fold) induced by ART alone, which were partially dysregulated by Mdr1a deficiency. Monotherapy and combination exert oppositely regulatory effects on the PI3K/AKT-mTOR pathway in a tissue-, Mdr1a genotype-, and parasite sensitivity/resistance-dependent manner (ranging from 1.52-fold to 84.00-fold). Specifically, CHR reversed ART-induced changes via PI3K/AKT protein inhibition (ranging from 1.20-fold to 63.00-fold), which was contingent on P-gp functionality. Finally, mitogen-activated protein kinase (MAPK) pathway was involved in the antagonistic regulation between ART alone and combination therapy in a P-gp-independent manner (ranging from 1.39-fold to 16.69-fold). Conclusions The efflux pump function of P-gp is probably not a critical factor in the mechanism by which CHR reverses ART resistance. Instead, CHR acts as a homeostasis stabilizer with dual functions: it disrupts Plasmodium berghei K173 resistance to ART driven by both ABC transporters and the heme-ROS/GSH axis, in which the non-transport function of P-gp on ART is involved. Graphical Abstract

Since Bitcoin was first conceived in 2008, blockchain technology has attracted a large amount of researchers’ attention. At the same time, it has also facilitated a variety of cybercrimes. For example, Ethereum frauds, due to the potential for huge profits, occur frequently and pose a serious threat to the financial security of the Ethereum network. To create healthy financial environments, methods for automatically detecting and identifying Ethereum frauds are urgently needed in Ethereum system governance. To this end, this paper proposes a new framework to detect fraudulent transactions in Ethereum by mining Ethereum transaction records. Specifically, we obtain Ethereum addresses with fraud/legitimate labels through Web crawlers and then construct a transaction network according to the public transaction ledger. Then, a transaction behavior-based network embedding algorithm is proposed to extract node features for subsequent fraudulent transaction identification. Finally, we adopt the Graph Convolutional Neural Network model (GCN) to classify addresses into legal and fraudulent addresses. The experimental results show that the fraudulent transaction detection system can achieve an accuracy of 96% on fraud/legitimate record classification, which proves the effectiveness of the framework in the detection of Ethereum fraudulent transactions.

AlN precipitates can be easily formed in Al-killed super-duplex stainless steel (SDSS), and may significantly deteriorate the mechanical properties of the steel. Therefore, three 2507 SDSS ingots containing 290 to 660 ppm Al were prepared, focusing on the characteristics of AlN precipitates, the solubility product of AlN and the mechanical properties. AlN precipitates were observed mainly in ferrite phase, and uniformly distributed over the phase. With increasing temperature from 1223 K to 1473 K, the area fraction of AlN precipitates approximately linearly decreases. However, further increasing temperature to 1573 K, the area fraction exhibits an obvious increasing tendency. Such a phenomenon is reported for the first time, and may be related to rapid variation of austenite and ferrite fractions over the temperature range. Based on a newly developed method, the expression of AlN solubility product as a function of temperature was obtained, and high reliability was demonstrated. The tensile/yield strength of 2507 SDSS are insensitive to Al content. However, the presence of 410 ppm and 660 ppm Al deteriorates the elongation at break and the impact energy, due to the formation of excessive AlN precipitates. This work provides useful guidance for the control of Al levels in 2507 SDSS.

Developmental dysplasia of the hip (DDH) is one of the most common diseases in children. Due to the experience-requiring medical image analysis work, online automatic diagnosis of DDH has intrigued the researchers. Traditional implementation of online diagnosis faces challenges with reliability and interpretability. In this paper, we establish an online diagnosis tool based on a multi-task hourglass network, which can accurately extract landmarks to detect the extent of hip dislocation and predict the age of the femoral head. Our method utilizes a multi-task hourglass network, which trains an encoder-decoder network to regress the landmarks and predict the developmental age for online DDH diagnosis. With the support of precise image analysis and fast GPU computing, our method can help overcome the shortage of medical resources and enable telehealth for DDH diagnosis. Applying this approach to a dataset of DDH X-ray images, we demonstrate 4.64 mean pixel error of landmark detection compared to the results of human experts. Moreover, we can improve the accuracy of the age prediction of femoral heads to 89%. Our online automatic diagnosis system has provided service to 112 patients, and the results demonstrate the effectiveness of our method.

Knowledge Graph Embedding (KGE) aims to learn representations of entities and relations of knowledge graph (KG). Recently Graph Neural Networks (GNNs) have gained great success on KGE, but for the reason behind it, most views simply attribute to the well learning of knowledge graph structure, which still remains a limited understanding of the internal mechanism. In this work, we first study a fundamental problem, i.e., what are the important factors for GNNs to help KGE. To investigate this problem, we discuss the core idea of current GNN models for KG, and propose a new assumption of relational homophily that connected nodes possess similar features after relation’s transforming, to explain why aggregating neighbors with relation can help KGE. Based on the model and empirical analyses, we then introduce a novel data-centric framework for applying GNNs to KGE called KSG-GNN . In KSG-GNN, we construct a new graph structure from KG named Knowledge Similarity Graph (KSG), where each node connects with its similar nodes as neighbors, and then we apply GNNs on this graph to perform KGE. Instead of following the relational homophily assumption in KG, KSG aligns with homogeneous graphs that can directly satisfy homophily assumption. Hence, any GNN developed on homogeneous graphs like GCN, GAT, GraphSAGE, etc., can be applied out-of-the-box as KSG-GNN without modification, which provides a more general and effective GNN paradigm. Finally, we conduct extensive experiments on two benchmark datasets, i.e., FB15k-237 and WN18RR, demonstrating the superior performance of KSG-GNN over multiple strong baselines. The source code is available at https://github.com/advancer99/WWWJ-KGE .

Semi-supervised learning (SSL) has attracted increasing attention in medical image segmentation, where the mainstream usually explores perturbation-based consistency as a regularization to leverage unlabelled data. However, unlike directly optimizing segmentation task objectives, consistency regularization is a compromise by incorporating invariance towards perturbations, and inevitably suffers from noise in self-predicted targets. The above issues result in a knowledge gap between supervised guidance and unsupervised regularization. To bridge the knowledge gap, this work proposes a meta-based semi-supervised segmentation framework with the exploitation of label hierarchy. Two main prominent components named Divide and Generalize, and Label Hierarchy, are built in this work. Concretely, rather than merging all knowledge indiscriminately, we dynamically divide consistency regularization from supervised guidance as different domains. Then, a domain generalization technique is introduced with a meta-based optimization objective which ensures the update on supervised guidance should generalize to the consistency regularization, thereby bridging the knowledge gap. Furthermore, to alleviate the negative impact of noise in self-predicted targets, we propose to distill the noisy pixel-level consistency by exploiting label hierarchy and extracting hierarchical consistencies. Comprehensive experiments on two public medical segmentation benchmarks demonstrate the superiority of our framework to other semi-supervised segmentation methods, with new state-of-the-art results.

Four sand-filled hydraulic fractures were created at a depth of 1.5 m, and the vicinities of the fractures were excavated and mapped in detail. All the fractures were shaped like slightly asymmetric saucers between 4.5 and 7.0 m across that were roughly flat-lying in their centers and curved upward to dip between 10° and 15° along their peripheries. Three different colors of sand were injected in sequence to trace the relative ages of the sand in the fracture. The first sand to be injected remained in the vicinity of the injection casing, whereas the last sand moved rapidly to the leading edge. Sand transport occurred through localized, channel-like pathways that extended from the injection casing and then branched into multiple paths as they approached the leading edge. At least four branching pathways of different ages were identified in one fracture, suggesting that this represents a fundamental mechanism of sand transport in these shallow fractures. A theoretical model was developed by adapting Franc2d, a code well-known in structural mechanics, to predict the propagation paths of curved hydraulic fractures at shallow depths. The model predicts fracture forms that are remarkably similar to those in field exposures when properties typical of field conditions are used.Key words: hydraulic fracture, field test, mechanics, modeling.

With the wide application of lithium ion battery in the energy storage system, Much attention had been paid to the state of health (SOH) evaluation research. In this paper, the research advance of SOH evaluation methods for lithium-ion battery was reviewed. Several main SOH evaluation methods, including defining method, internal resistance method, AC impedance, voltage curve method, model method and some new methods, were discussed in detail. And finally the research direction of further progress in this area was proposed.