The impressive performance of Large Language Model (LLM) has prompted researchers to develop Multi-modal LLM (MLLM), which has shown great potential for various multi-modal tasks. However, current MLLM often struggles to effectively address fine-grained multi-modal challenges. We argue that this limitation is closely linked to the models’ visual grounding capabilities. The restricted spatial awareness and perceptual acuity of visual encoders frequently lead to interference from irrelevant background information in images, causing the models to overlook subtle but crucial details. As a result, achieving fine-grained regional visual comprehension becomes difficult. In this paper, we break down multi-modal understanding into two stages, from Coarse to Fine (CoF). In the first stage, we prompt the MLLM to locate the approximate area of the answer. In the second stage, we further enhance the model’s focus on relevant areas within the image through visual prompt engineering, adjusting attention weights of pertinent regions. This, in turn, improves both visual grounding and overall performance in downstream tasks. Our experiments show that this approach significantly boosts the performance of baseline models, demonstrating notable generalization and effectiveness. Our CoF approach is available online at https://github.com/Gavin001201/CoF.

Despite the significant success of Large Vision-Language models(LVLMs), these models still suffer hallucinations when describing images, generating answers that include non-factual objects. It is reported that these models tend to overfocus on certain irrelevant image tokens that do not contain critical information for answering the question and distort the output. To address this, we propose an Instruction-Aligned Visual Attention(IAVA) approach, which identifies irrelevant tokens by comparing changes in attention weights under two different instructions. By applying contrastive decoding, we dynamically adjust the logits generated from original image tokens and irrelevant image tokens, reducing the model’s over-attention to irrelevant information. The experimental results demonstrate that IAVA consistently outperforms existing decoding techniques on benchmarks such as MME, POPE, and TextVQA in mitigating object hallucinations. Our IAVA approach is available online at https://github.com/Lee-lab558/IAVA.

Privacy and security concerns are becoming increasingly critical for recommender systems, as model extraction attack provides an effective way to probe system robustness by replicating the model’s recommendation logic — potentially exposing sensitive user preferences and proprietary algorithmic knowledge. Despite the promising performance of existing model extraction methods, they still face two key challenges: unrealistic assumptions on the requirement of accessible member or surrogate data and generalization problem where surrogate model architecture constraints lead to overfitting on generated data. To tackle these challenges, in this paper, we first thoroughly analyze how the architecture of surrogate models influences extraction attack performance, highlighting the superior effectiveness of the graph convolution architecture. Based on this, we propose a novel Data-free Black-box Graph convolution-based Recommender Model Extraction method, dubbed DBGRME. Specifically, DBGRME contains: (1) an interaction generator to alleviate the need for member data requirements in a data-free scenario; and (2) a generalization-aware graph convolution-based surrogate model to capture diverse and complex recommender interaction patterns for mitigating the overfitting issue. Experimental results on various datasets and victim models demonstrate the superiority of our attack in data-free scenarios (e.g., surpassing PTQ data-require methods with 17.4% improvement on LightGCN). Code is available: \url{https://github.com/Vencent-Won/DBGRME.git}.

As the Internet rapidly expands, the increasing complexity and diversity of network activities pose significant challenges to effective network governance and security regulation. Network traffic, which serves as a crucial data carrier of network activities, has become indispensable in this process. Network traffic detection aims to monitor, analyze, and evaluate the data flows transmitted across the network to ensure network security and optimize performance. However, existing network traffic detection methods generally suffer from several limitations: 1) a narrow focus on characterizing traffic features from a single perspective; 2) insufficient exploration of discriminative features for different traffic; 3) poor generalization to different traffic scenarios. To address these issues, we propose a multi-view correlation-aware framework namedFlowIDfor network traffic detection.FlowIDcaptures multi-view traffic features via temporal and interaction awareness, while a hypergraph encoder further explores higher-order relationships between flows. To overcome the challenges of data imbalance and label scarcity, we design a dual-contrastive proxy task, enhancing the framework’s ability to differentiate between various traffic flows through flow-to-flow and group-to-group contrast. Extensive experiments on five real-world datasets demonstrate thatFlowIDsignificantly outperforms existing methods in accuracy, robustness, and generalization across diverse network scenarios, particularly in detecting malicious traffic.

Abstract Link prediction, which aims to infer missing edges or predict future edges based on currently observed graph connections, has emerged as a powerful technique for diverse applications such as recommendation, relation completion, etc. While there is rich literature on link prediction based on node representation learning, direct link embedding is relatively less studied and less understood. One common practice in previous work characterizes a link by manipulate the embeddings of its incident node pairs, which is not capable of capturing effective link features. Moreover, common link prediction methods such as random walks and graph auto-encoder usually rely on full-graph training, suffering from poor scalability and high resource consumption on large-scale graphs. In this paper, we propose Inductive Subgraph Embedding for Link Prediciton (SE4LP) — an end-to-end scalable representation learning framework for link prediction, which utilizes the strong correlation between central links and their neighborhood subgraphs to characterize links. We sample the “link-centric induced subgraphs” as input, with a subgraph-level contrastive discrimination as pretext task, to learn the intrinsic and structural link features via subgraph classification. Extensive experiments on five datasets demonstrate that SE4LP has significant superiority in link prediction in terms of performance and scalability, when compared with state-of-the-art methods. Moreover, further analysis demonstrate that introducing self-supervision in link prediction can significantly reduce the dependence on training data and improve the generalization and scalability of model.

Graph neural networks (GNNs) have achieved remarkable advances in graph-oriented tasks. However, real-world graphs invariably contain a certain proportion of heterophilous nodes, challenging the homophily assumption of traditional GNNs and hindering their performance. Most existing studies continue to design generic models with shared weights between heterophilous and homophilous nodes. Despite the incorporation of high-order messages or multi-channel architectures, these efforts often fall short. A minority of studies attempt to train different node groups separately but suffer from inappropriate separation metrics and low efficiency. In this paper, we first propose a new metric, termed Neighborhood Confusion (NC), to facilitate a more reliable separation of nodes. We observe that node groups with different levels of NC values exhibit certain differences in intra-group accuracy and visualized embeddings. These pave the way for Neighborhood Confusion-guided Graph Convolutional Network (NCGCN), in which nodes are grouped by their NC values and accept intra-group weight sharing and message passing. Extensive experiments on both homophilous and heterophilous benchmarks demonstrate that our framework can effectively separate nodes and yield significant performance improvement compared to the latest methods.

The rampant fraudulent activities on Ethereum hinder the healthy development of the blockchain ecosystem, necessitating the reinforcement of regulations. However, multiple imbalances involving account interaction frequencies and interaction types in the Ethereum transaction environment pose significant challenges to data mining-based fraud detection research. To address this, we first propose the concept of meta-interactions to refine interaction behaviors in Ethereum, and based on this, we present a dual self-supervision enhanced Ethereum fraud detection framework, named Meta-IFD. This framework initially introduces a generative self-supervision mechanism to augment the interaction features of accounts, followed by a contrastive self-supervision mechanism to differentiate various behavior patterns, and ultimately characterizes the behavioral representations of accounts and mines potential fraud risks through multi-view interaction feature learning. Extensive experiments on real Ethereum datasets demonstrate the effectiveness and superiority of our framework in detecting common Ethereum fraud behaviors such as Ponzi schemes and phishing scams. Additionally, the generative module can effectively alleviate the interaction distribution imbalance in Ethereum data, while the contrastive module significantly enhances the framework’s ability to distinguish different behavior patterns. The source code will be available inhttps://github.com/GISec-Team/Meta-IFD.

Molecular property prediction faces the challenge of limited labeled data as it necessitates a series of specialized experiments to annotate target molecules. Data augmentation techniques can effectively address the issue of data scarcity. In recent years, Mixup has achieved significant success in traditional domains such as image processing. However, its application in molecular property prediction is relatively limited due to the irregular, non-Euclidean nature of graphs and the fact that minor variations in molecular structures can lead to alterations in their properties. To address these challenges, we propose a novel data augmentation method called Mix-Key tailored for molecular property prediction. Mix-Key aims to capture crucial features of molecular graphs, focusing separately on the molecular scaffolds and functional groups. By generating isomers that are relatively invariant to the scaffolds or functional groups, we effectively preserve the core information of molecules. Additionally, to capture interactive information between the scaffolds and functional groups while ensuring correlation between the original and augmented graphs, we introduce molecular fingerprint similarity and node similarity. Through these steps, Mix-Key determines the mixup ratio between the original graph and two isomers, thus generating more informative augmented molecular graphs. We extensively validate our approach on molecular datasets of different scales with several Graph Neural Network architectures. The results demonstrate that Mix-Key consistently outperforms other data augmentation methods in enhancing molecular property prediction on several datasets.

Graph neural networks (GNNs) have achieved remarkable success in various real-world applications. However, recent studies highlight the vulnerability of GNNs to malicious perturbations. Previous adversaries primarily focus on graph modifications or node injections to existing graphs, yielding promising results but with notable limitations. Graph modification attack (GMA) requires manipulation of the original graph, which is often impractical, while graph injection attack (GIA) necessitates training a surrogate model in the black-box setting, leading to significant performance degradation due to divergence between the surrogate architecture and the actual victim model. Furthermore, most methods concentrate on a single attack goal and lack a generalizable adversary to develop distinct attack strategies for diverse goals, thus limiting precise control over victim model behavior in real-world scenarios. To address these issues, we present a gradient-free generalizable adversary that injects a single malicious node to manipulate the classification result of a target node in the black-box evasion setting. Specifically, we model the single-node injection label specificity attack as a Markov decision process (MDP) and propose gradient-free generalizable single node injection attack, namely G2-SNIA, a reinforcement learning framework employing proximal policy optimization (PPO). By directly querying the victim model, G2-SNIA learns patterns from exploration to achieve diverse attack goals with extremely limited attack budgets. Through comprehensive experiments over three acknowledged benchmark datasets and four prominent GNNs in the most challenging and realistic scenario, we demonstrate the superior performance of our proposed G2-SNIA over the existing state-of-the-art baselines. Moreover, by comparing G2-SNIA with multiple white-box evasion baselines, we confirm its capacity to generate solutions comparable to those of the best adversaries.

Critical node detection (CND) is commonly used to detect nodes with a high impact on network robustness. It has been widely used in disease propagation, social networks, communications, and other fields. As a nondeterministic polynomial-time (NP)-complete problem, the efficiency of solving CND severely limits the scale of the available network. Fortunately, the evolutionary algorithm (EA) is an effective method to solve this problem. However, although EA improves the global search capability of the algorithm by preserving gene diversity, it also introduces many inferior genes, thus expanding the candidate solution space, reducing the search efficiency, and making it difficult to apply the pruning algorithm directly to its solution space. Hence, indirectly reducing the solution space of EA by deleting inferior genes is a feasible pruning method; however, the interaction of multiple genes affects the quality of CND solutions, making it a challenge to pick out inferior individual genes. Therefore, this work proposes a multipopulation synergistic gene screening algorithm based on the parallelism of EA and combined with Ensemble learning for identifying low-quality genes and removing them as a way of pruning the solution space of the algorithm and improving the search efficiency. The algorithm encodes all nodes in the graph as the gene pool of EA and treats a single population as a weak learner to screen the dominant genes in the gene pool and achieve fast pruning of EA’s solution space by integrating the dominant individuals in multiple populations. In this work, the experiments demonstrate the effectiveness of the proposed method and analyze the effect of different network structures on the algorithm.

With the rapid development of artificial intelligence, a number of machine learning algorithms, such as graph neural networks (GNNs), have been proposed to facilitate network analysis or graph data mining. Although effective, recent studies show that these advanced methods may suffer from adversarial attacks, i.e., they may lose effectiveness when only a small fraction of links are unexpectedly changed. This article investigates three well-known adversarial attack methods, i.e., Nettack, Meta Attack, and GradArgmax. It is found that different attack methods have their specific attack preferences on changing the target network structures. Such attack patterns are further verified by experimental results on some real-world networks, revealing that, generally, the top-4 most important network attributes on detecting adversarial samples suffice to explain the preference of an attack method. Based on these findings, the network attributes are utilized to design machine learning models for adversarial sample detection and attack method recognition with outstanding performance.

The recent flourish of deep learning on various tasks is largely accredited to the rich and high-quality labeled data. Nonetheless, collecting sufficient labeled samples is not very practical for many real applications. Few-shot Learning (FSL) provides a promising solution that allows a model to learn the concept of novel classes with a few labeled samples. However, many existing FSL methods are only designed for computer vision tasks and are not suitable for radio signal recognition. This paper calls for a radically different approach to FSL: in contrast to developing a new FSL model, we should focus on transforming the radio signal to be better processed by the state-of-the-art (SOTA) FSL model. We propose Modulated Signal Pre-transformation (MSP), a parameterized radio signal transformation framework that encourages the signals having the same semantics to have similar representations. MSP currently adapts to various SOTA FSL models for signal modulation recognition and can support the mainstream deep learning backbone. Evaluation results show that MSP improves the performance gains for many SOTA FSL models while maintaining flexibility.

Graph neural network (GNN) has achieved great success on graph representation learning. Challenged by large-scale private data collected from user side, GNN may not be able to reflect the excellent performance, without rich features and complete adjacent relationships. Addressing the problem, vertical federated learning (VFL) is proposed to implement local data protection through training a global model collaboratively. Consequently, for graph-structured data, it is a natural idea to construct a GNN-based VFL (GVFL) framework. However, GNN has been proven vulnerable to adversarial attacks. Whether the vulnerability will be brought into the GVFL has not been studied. This is the first study of adversarial attacks on GVFL. A novel adversarial attack method is proposed, named Graph-Fraudster. It generates adversarial perturbations based on the noise-added global node embeddings via the privacy leakage and the gradient of pairwise node. Specifically, first, Graph-Fraudster steals the global node embeddings and sets up a shadow model of the server for the attack generator. Second, noise is added into node embeddings to confuse the shadow model. Finally, the gradient of pairwise node is used to generate attacks with the guidance of noise-added node embeddings. Extensive experiments on five benchmark datasets demonstrate that Graph-Fraudster achieves the state-of-the-art attack performance compared with baselines in different GNN based GVFLs. Furthermore, Graph-Fraudster can remain a threat to GVFL even if two possible defense mechanisms are applied. In addition, some suggestions are put forward for the future work to improve the robustness of GVFL. The code and datasets can be downloaded athttps://github.com/hgh0545/Graph-Fraudster.

Critical nodes detection (CND) focuses on identifying the nodes that significantly impact the network’s robustness and is applied in various fields such as power grids, communication networks, and disease spreading. However, detecting the critical nodes is a challenging nondeterministic polynomial time complete (NP-complete) problem. One possible solution is using the evolutionary algorithm which has a high global search capability. However, the existing evolutionary algorithms for CND only focus on independent nodes, ignoring the underlying relationship among the nodes. Thus, in this work, we proposed a new topology-combined differential evolution framework called TDE to explore the possibility of improving the performance by fusing topology information, which designs individual genotypes through node degree, and new mutation and decoding-based selection operators are designed for these genotypes to use topology information effectively. The experiments on synthetic and real networks show that it is feasible to improve the search capability of the algorithm by fusing node degree information.

Community detection, which focuses on clustering vertex interactions, plays a significant role in network analysis. However, it also faces numerous challenges like missing data and adversarial attack. How to further improve the performance and robustness of community detection for real-world networks has raised great concerns. In this paper, we explore robust community detection by enhancing network structure, with two generic algorithms presented: one is named robust community detection via genetic algorithm (RobustECDGA), in which the modularity and the number of clusters are combined in a fitness function to find the optimal structure enhancement scheme; the other is called robust community detection via similarity ensemble (RobustECD-SE), integrating multiple information of community structures captured by various vertex similarities, which scales well on large-scale networks. Comprehensive experiments on real-world networks demonstrate, by comparing with two traditional enhancement strategies, that the new methods help six representative community detection algorithms achieve more significant performance improvement. Moreover, experiments on the corresponding adversarial networks indicate that the new methods could also optimize the network structure to a certain extent, achieving stronger robustness against adversarial attack. The source code of this paper is released on https://github.com/jjzhou012/robustECD release.

Location-Based Social Networks have been widely studied in recent years; new approaches constantly developed to solve individuals’ trajectory prediction tasks. However, most of these methods require sufficient data to learn individual features, which is not always satisfied in real situations, especially for online data. The digital data on human behavior typically follows a power-law distribution, indicating that only a few people have rich activities recorded while most people’s behavioral data are limited. In order to overcome this hurdle, our work constructs the user behavior proximity network (UBPN) and proposes a new walking strategy based on this network that extracts the hidden information from the social contacts to substitute the unobserved behavioral information of an individual. Specifically, our proposed walking strategy has two walking paths, accounting for the temporal and social information on the ego users’ and their alters’ mobility activities. This walking strategy is model-agnostic and can be integrated with many existing walk-based deep learning methods. Our work applies the methods on two real-world datasets with rich spatiotemporal information and shows that the performances of the existing prediction methods improve significantly by integrating the proposed walking strategy.

The adversarial attack methods based on gradient information can adequately find the perturbations, that is, the combinations of rewired links, thereby reducing the effectiveness of the deep learning model-based graph embedding algorithms, but it is also easy to fall into a local optimum. Therefore, this article proposes a momentum gradient attack (MGA) against the graph convolutional network (GCN) model, which can achieve more aggressive attacks with fewer rewiring links. Compared with directly updating the original network using gradient information, integrating the momentum term into the iterative process can stabilize the updating direction, which makes the model jump out of poor local optimum and enhances the method with stronger transferability. Experiments on node classification and community detection methods based on three well-known network embedding algorithms show that MGA has a better attack effect and transferability.

In social networks, by removing some target-sensitive links, privacy protection might be achieved. However, some hidden links can still be re-observed by link prediction methods on observable networks. In this paper, the conventional link prediction method named Resource Allocation Index (RA) is adopted for privacy attacks. Several defense methods are proposed, including heuristic and evolutionary approaches, to protect targeted links from RA attack. In particular, incremental computation is proposed for accelerating the calculation of fitness in evolutionary approaches. This is the first time to study privacy protection for targeted links against similarity based link prediction attacks. Some links are randomly selected from original network as targeted links for experimentation. The experimental results on nine real-world networks demonstrate the superiority of the evolutionary perturbations, especially EDA, for defending against RA attack. Moreover, experimental results show that the proposed perturbation generated by EDA is transferable and can even defend against other link prediction attacks which are based on high order similarity between pairwise nodes, although it is designed to prevent RA attack.

Community detection plays an important role in social networks, since it can help to naturally divide the network into smaller parts so as to simplify network analysis. However, on the other hand, it arises the concern that individual information may be overmined, and the concept community deception has been proposed to protect individual privacy on social networks. Here, we introduce and formalize the problem of community detection attack and develop efficient strategies to attack community detection algorithms by rewiring a small number of connections, leading to privacy protection. In particular, we first give two heuristic attack strategies, i.e., Community Detection Attack (CDA) and Degree Based Attack (DBA), as baselines, utilizing the information of detected community structure and node degree, respectively. Then, we propose an attack strategy called “genetic algorithm (GA)-based Q-Attack,” where the modularity Q is used to design the fitness function. We launch community detection attack based on the above three strategies against six community detection algorithms on several social networks. By comparison, our Q-Attack method achieves much better attack effects than CDA and DBA, in terms of the larger reduction of both modularity Q and normalized mutual information (NMI). In addition, we further take transferability tests and find that adversarial networks obtained by Q-Attack on a specific community detection algorithm also show considerable attack effects while generalized to other algorithms.

Recommender systems are becoming more and more important in our daily lives. However, traditional recommendation methods are challenged by data sparsity and efficiency, as the numbers of users, items, and interactions between the two in many real-world applications increase fast. In this paper, we propose a novel clustering recommender system based on node2vec technology and rich information network, namely, N2VSCDNNR, to solve these challenges. In particular, we use a bipartite network to construct the user-item network and represent the interactions among users (or items) by the corresponding one-mode projection network. In order to alleviate the data sparsity problem, we enrich the network structure according to user and item categories, and construct the one-mode projection category network. Then, considering the data sparsity problem in the network, we employ node2vec to capture the complex latent relationships among users (or items) from the corresponding one-mode projection category network. Moreover, considering the dependence on parameter settings and information loss problem in clustering methods, we use a novel spectral clustering method, which is based on dynamic nearest-neighbors (DNNs) and a novel automatically determining cluster number (ADCN) method that determines the cluster centers based on the normal distribution method, to cluster the users and items separately. After clustering, we propose the two-phase personalized recommendation to realize the personalized recommendation of items for each user. A series of experiments validate the outstanding performance of our N2VSCDNNR over several advanced embedding and side information based recommendation algorithms. Meanwhile, N2VSCDNNR seems to have lower time complexity than the baseline methods in online recommendations, indicating its potential to be widely applied in large-scale systems.

This paper describes a kind of replacement mechanism which improves the generalization ability of Variable Size Genetic Network Programming (GNPvs). GNPvs is an extension of Genetic Network Programming (GNP), which has changeable number of nodes. Inspired by the theory of Evolution by Gene Duplication, the non-frequently used nodes are replaced with the frequently used nodes in the proposed method, which can make the individual survive under the selection pressure as usual, but eventually might accumulate mutations that produce new features of individuals for adapting new environments. The effectiveness of the proposed method is verified by comparing with the performance of GNPvs and GNP on a well-known dynamic multi-agent test bed - Tile world.

In order to extract class association rules more effectively when dealing with large attributes set, Genetic Network Programming (GNP) based class association rule mining with Attributes Importance has been proposed in this paper. The main difference between the proposed method and the conventional GNP-based class association rule mining is that Attributes Importance is introduced to affect the attributes selection and genetic operations during the GNP evolution process. The comparison has been carried out by applying the proposed method and the conventional GNP-based class association rule mining to the rules extraction with regard to the interested products on the Internet shop for different customers. The simulation results shows that the efficiency of rules extraction is improved greatly by adopting the proposed method.

In this paper, a distributed dynamic traffic management model has been proposed to guide the vehicles, in order to minimize the computation time, make full use of real time traffic information and consequently improve the efficiency of the traffic system. For making the model work, we proposed a new dynamic route determination method, in which Q value-based Dynamic Programming and Sarsa Learning are combined to calculate the approximate optimal traveling time from each section to the destinations in the road networks. The proposed traffic management model is applied to the large scale microscopic simulator SOUND/4U based on the real world road network of Kurosaki, Kitakyushu in Japan. The simulation results show that the proposed method could reduce the traffic congestion and improve the efficiency of the traffic system effectively compared with the conventional method in the real world road network.

In this paper, a dynamic traffic management model has been proposed to alleviate the traffic congestion and improve the efficiency of the traffic systems in global perspective. The proposed traffic management model is applied to the large scale microscopic simulator SOUND/4U based on the real world road network of Kurosaki, Kitakyushu in Japan. All the vehicles in the simulator follow the direction from the route guidance of the dynamic traffic management model, in which the extended Q value-based Dynamic Programming with Boltzmann Distribution and the time-varying traffic information are used to generate the routes from the origins to destinations. The simulation results show that the proposed Q value-based Dynamic Programming with Boltzmann Distribution could reduce the traffic congestion and improve the efficiency of the whole traffic system effectively compared with the greedy method in the real world road network.

In order to alleviate the congestion in modern metropolises with over crowded traffics and improve the efficiency of Intelligent Transportation Systems, three temperature parameter control methods of Q value-based Dynamic Programming with Boltzmann Distribution have been proposed in this paper. The simulation result shows that each method has its own areas of expertise depending on its features and all of the methods could improve the efficiency of the traffic system comparing with the conventional Greedy Method.

In this paper, we propose a heuristic method trying to improve the efficiency of traffic systems in the global perspective, where the optimal traveling time for each origin-destination (OD) pair is calculated by extended Q value-based dynamic programming and the global optimum routes are produced by adjusting the temperature parameter in Boltzmann distribution. The key point is that the temperature parameter for each section is not identical, but constantly changing with the traffic of the section, which enables the diversified routing strategy depending on the latest traffics. In addition, the simulation results show that comparing with the greedy strategy and constant temperature parameter strategy, the proposed method, i.e., temperature parameter control strategy of the Q value-based dynamic programming with Boltzmann distribution, could reduce the traffic congestion effectively and minimize the negative impact of the information update interval by adopting suitable temperature parameter control strategy.