In this paper, we investigate the discovery of temporal anomalous subgraphs in large-scale financial networks, aiming to identify abnormal transaction behaviors among users over time. This task is crucial for the real-time detection of transaction anomalies in financial networks, such as money laundering and trading fraud. However, it poses significant challenges due to the diverse distribution of transactions, the dynamic nature of temporal networks, and the absence of theoretical foundation. To tackle these challenges, we introduce a novel Temporal Anomalous Subgraph Discovery (TempASD) algorithm with theoretical analysis. First, we propose a temporal candidate detection module that quickly pinpoints abnormal candidates by detecting anomalies in both the temporal structure and transaction distribution. Then, we introduce a carefully crafted reinforcement-learning-based refiner to optimize these candidates toward the most abnormal directions. We conducted extensive evaluations against thirteen advanced competitors. TempASD achieves an average improvement of 7x in abnormal degree compared to the state-of-the-art and is efficient in large-scale dynamic financial networks.
论文检索
输入标题、作者或关键词，从 844 篇学术成果中精准定位
Visual geo-localization demands in-depth knowledge and advanced reasoning skills to associate images with precise real-world geo-graphic locations. Existing image database retrieval methods are limited by the impracticality of storing sufficient visual records of global landmarks. Recently, Large Vision-Language Models (LVLMs) have demonstrated the capability of geo-localization through Visual Question Answering (VQA), enabling a solution that does not require external geo-tagged image records. However, the performance of a single LVLM is still limited by its intrinsic knowledge and reasoning capabilities. To address these challenges, we introduce smileGeo, a novel visual geo-localization framework that leverages multiple Internet-enabled LVLM agents operating within an agent-based architecture. By facilitating inter-agent communication, smileGeo integrates the inherent knowledge of these agents with additional retrieved information, enhancing the ability to effectively localize images. Furthermore, our framework incorporates a dynamic learning strategy that optimizes agent communication, reducing redundant interactions and enhancing overall system efficiency. To validate the effectiveness of the proposed framework, we conducted experiments on three different datasets, and the results show that our approach significantly outperforms current state-of-the-art methods. The source code is available at https://github.com/Applied-Machine-Learning-Lab/smileGeo.
Dynamic graphs offer more precise modeling of real-world applications compared to static graphs. Therefore, learning on dynamic graphs has garnered significant research attention in recent years. Unfortunately, the current approaches for learning dynamic graphs remain inadequate in capturing global spatial-temporal dependency. This issue arises from capturing biased spatial and temporal dependencies, thereby weakening the coupling between spatial and temporal dimensions. To overcome it, we propose a Self-Reliant Global Dependency Apprehending Framework on Dynamic Graphs, namely SGD-DyG. Specifically, we first design a frequency self-enhanced learning module that examines the global inherent interactions of the node features hidden in the frequency domain. Furthermore, we propose a global-local-mixed self-supervised learning module that maximizes spatial-temporal mutual information between local node and global graph embeddings. Extensive experiments on seven real-world dynamic graph datasets validate that the proposed SGD-DyG consistently exceeds state-of-the-art models.
In automated decision-making, it is desirable that outputs of decision-makers be robust to slight perturbations in their inputs, a property that may be called input-output robustness. Input-output robustness appears in various different forms in the literature, such as robustness of AI models to adversarial or semantic perturbations and individual fairness of AI models that make decisions about humans. We propose runtime monitoring of input-output robustness of deployed, black-box AI models, where the goal is to design monitors that would observe one long execution sequence of the model, and would raise an alarm whenever it is detected that two similar inputs from the past led to dissimilar outputs. This way, monitoring will complement existing offline ''robustification'' approaches to increase the trustworthiness of AI decision-makers. We show that the monitoring problem can be cast as the fixed-radius nearest neighbor (FRNN) search problem, which, despite being well-studied, lacks suitable online solutions. We present our tool Clemont, which offers a number of lightweight monitors, some of which use upgraded online variants of existing FRNN algorithms, and one uses a novel algorithm based on binary decision diagrams--a data-structure commonly used in software and hardware verification. We have also developed an efficient parallelization technique that can substantially cut down the computation time of monitors for which the distance between input-output pairs is measured using the L∞norm. Using standard benchmarks from the literature of adversarial and semantic robustness and individual fairness, we perform a comparative study of different monitors in Clemont, and demonstrate their effectiveness in correctly detecting robustness violations at runtime.
User personas play a crucial role in shaping shopping behavior, making persona identification an essential task for e-commerce platforms to personalize user experiences. However, this problem remains highly challenging due to several key factors: (1) users often exhibit multiple personas simultaneously (e.g., a fashion enthusiast who is also a sports lover)(2) labeled data is scarce, as annotation requires extensive human supervision(3) real-world user-product interaction data is inherently noisy, as accounts may be shared among multiple individuals, complicating persona assessment, and (4) user-product interactions form a dynamic, heterogeneous bipartite graph where product features are diverse, and personas evolve due to shifts in interests, seasons, and external events. In this work, we study the problem of persona identification on an extensive real-world user-product interaction dataset spanning six months (August 2023 to January 2024), capturing user behavior influenced by real-world factors such as seasonal changes, festivals (e.g., Christmas, New Year), and major sales events. To model persona identification in this evolving interaction graph, we reformulate the multi-label node classification task as a link prediction problem, enabling a structured decoupling of user and persona representations. To this end, we propose TRIPER, a novel TRIpartite graph neural network specifically designed to enhance multi-label PERsona classification along with in-context inference capabilities. Extensive evaluations on our real-world dataset demonstrate that TRIPER achieves high predictive accuracy in user persona identification, exhibits strong generalization over time, and effectively learns from limited labeled data, outperforming state-of-the-art baselines.
In this paper, we reveal that existing Differentially Private Graph Neural Networks (DP-GNNs) are not effective against Graph Reconstruction Attack (GRA). We further attribute the ineffectiveness of existing DP-GNNs against GRA to their unstructured perturbation mechanism, which only induces unidirectional shift in the embedding similarity distribution. Specifically, this perturbation mechanism tends to decrease the embedding similarity of all node pairs without significantly disrupting the relative ranking, thus allowing GRA to still reconstruct the original graph structure by leveraging the relative ranking of similarities. To address this, we propose a novel Differentially Private Graph Neural Network based on Structured Perturbation (GRASP). Specifically, we observe that independent noise tends to decrease the embedding similarity, while identical noise tends to increase it. By integrating these two types of noise using a Bernoulli technique, we introduce a simple yet effective structured perturbation mechanism, which promotes bidirectional shift in the embedding similarity distribution, thereby effectively disrupting the relative ranking and defending against GRA. Extensive experiments on eight benchmark datasets demonstrate that GRASP effectively defends against GRA. Furthermore, GRASP achieves a superior privacy-utility trade-off compared to existing graph structure protection methods. The implementation of GRASP is available at https://github.com/ZhiyuZone/GRASP/.
Graph Alignment (GA) is an NP-hard combinatorial challenge. The existing methods usually work on small-scale graphs because their computational complexity grows as the square of the number of nodes. Recent techniques relied on alignment labels to divide the large source and target graphs into small ones respectively, which are then aligned separately. However, the alignment labels in expansive real-world graphs are often scarce, which makes GA even more challenging. To address these issues, we propose a novel self-supervised learning framework that is able to work on the graphs of million nodes without any alignment labels, where the existing methods usually fail to output an answer within a reasonable time. Our method also falls in the divide-and-conquer paradigm. We do not cluster the source and target graph separately by the existing graph clustering algorithms, because it is difficult to pair and then align the cluster-induced source and target subgraphs in the absence of alignment labels. We devise an Alignment-Aware Clustering (AAC) method to partition the source and target nodes jointly and keep the ought-to-be-aligned pairs of nodes within the same cluster as much as possible. Furthermore, we develop a Topology-Aware Repartition (TAR) that leverages pseudo-alignment labels as anchors for re-clustering, preserving detailed structural information within clusters. This enables graph neural networks to effectively enhance the graph representation learning for node alignments which are self-supervised by a graph-matching solver. Extensive experiments have demonstrated that our method greatly reduces computational demands and sustains high alignment accuracies in large-scale graph applications.
Different from existing vertical federated learning (VFL), zero-shot VFL (ZVFL) is an under-explored scenario where test classes are absent from partial parties' training sets. In extreme cases, some test classes even have no training samples for all parties. Traditionally, existing zero-shot methods require abundant seen-class samples for effective knowledge transfer to recognize unseen classes. However, both the limited aligned samples and different seen classes pose several unique challenges to ZVFL. The primary challenge lies in the seen-to-unseen transfer insufficiency, as the scarcity of aligned samples and diverse seen-class distributions across parties severely limits the model's capability to learn discriminative features that can generalize to unseen classes. Moreover, the multi-party bias inconsistency arises as different parties tend to be biased towards their own seen classes during prediction, leading to skewed classification results at the active party. To address these challenges, we propose FEZE, an alignment-flexible zero-shot vertical federated learning framework. Specifically, we introduce a relation learning network to capture class-feature relationships between class labels and feature representations across heterogeneous feature spaces, enabling unseen class recognition through relationship inference. Additionally, we design a meta-relation learning mechanism that leverages diverse class-feature patterns to tackle the insufficient feature generalization from limited seen-class samples. Finally, we propose an alignment-flexible adaptive aggregation strategy that achieves adaptively aggregation based on inconsistent prediction spaces with arbitrary number of aligned samples. Theoretical analysis proves that FEZE can achieve a convergence rate of O(1/T). In the most challenging zero-shot scenario without aligned samples, FEZE surpasses state-of-the-art baselines by an average of 7.47% across three datasets.
Homomorphism is an important structure-preserving mapping between graphs. Given a graph G and a pattern Q, the subgraph homomorphism problem is to find a mapping φ from Q to G such that adjacent vertices of Q are mapped to adjacent vertices in G. Unlike the subgraph isomorphic mapping that is injective, homomorphism allows multiple vertices in Q to map to the same vertex in G, increasing complexity. We develop HFrame, the first GNN-based framework for subgraph homomorphism, by combining algorithms and machine learning. We show that HFrame is more expressive than the vanilla GNN, i.e., HFrame can distinguish more graph pairs (Q, G) such that Q is not homomorphic to G. Moreover, we provide a generalization error bound for HFrame. Using real-life and synthetic graphs, we show that HFrame is up to 101.91× faster than exact matching algorithms, and its average accuracy can reach 0.962.
Converting webpage designs into code (design-to-code) plays a vital role in User Interface (UI) development for front-end developers, bridging the gap between visual design and functional implementation. While recent Multimodal Large Language Models (MLLMs) have shown significant potential in design-to-code tasks, they often fail to accurately preserve the layout during code generation. To this end, we draw inspiration from the Chain-of-Thought (CoT) reasoning in human cognition and propose LaTCoder, a novel approach that enhances layout preservation in webpage design during code generation with Layout-as-Thought (LaT). Specifically, we first introduce a simple yet efficient algorithm to divide the webpage design into image blocks. Next, we prompt MLLMs using a CoT-based approach to generate code for each block. Finally, we apply two assembly strategies-absolute positioning and an MLLM-based method-followed by dynamic selection to determine the optimal output. We evaluate the effectiveness of LaTCoder using multiple backbone MLLMs (i.e., DeepSeek-VL2, Gemini, and GPT-4o) on both a public benchmark and a newly introduced, more challenging benchmark (CC-HARD) that features complex layouts. The experimental results on automatic metrics demonstrate significant improvements. Specifically, TreeBLEU scores increased by 66.67% and MAE decreased by 38% when using DeepSeek-VL2, compared to direct prompting. Moreover, the human preference evaluation results indicate that annotators favor the webpages generated by LaTCoder in over 60% of cases, providing strong evidence of the effectiveness of our method.
Graph learning has been crucial to many real-world tasks, but they are often studied with a closed-world assumption, with all possible labels of data known a priori. To enable effective graph learning in an open and noisy environment, it is critical to inform the model users when the model makes a wrong prediction to in-distribution data of a known class, i.e., misclassification detection or when the model encounters out-of-distribution from novel classes, i.e., out-of-distribution detection. This paper introduces Evidential Reasoning Network (EVINET), a framework that addresses these two challenges by integrating Beta embedding within a subjective logic framework. EVINET includes two key modules: Dissonance Reasoning for misclassification detection and Vacuity Reasoning for out-of-distribution detection. Extensive experiments demonstrate that EVINET outperforms state-of-the-art methods across multiple metrics in the tasks of in-distribution classification, misclassification detection, and out-of-distribution detection. EVINET demonstrates the necessity of uncertainty estimation and logical reasoning for misclassification detection and out-of-distribution detection and paves the way for open-world graph learning. Our code and data are available at https://github.com/SSSKJ/EviNET.
Ranking systems based on pairwise comparisons are fundamental in decision-making applications, yet fairness concerns remain largely unaddressed, particularly in active ranking frameworks. We propose a novel fairness-aware active ranking approach that adaptively queries pairwise preferences to construct rankings that are both probably approximately correct (PAC) and fair. We propose a flexible cascaded norm-based objective that balances error distribution within and across socially salient groups, providing a unified framework to mitigate systemic disparities. Adopting our objective function allows us to explore fundamental fairness concepts like equal or proportionate errors within a unified framework. We develop both group-blind and group-aware algorithms and derive their sample complexity bounds. Empirical evaluations on real-world datasets, including COMPAS and German Credit, demonstrate the efficiency of our approach, reducing sample complexity while achieving fairer rankings. Our findings offer theoretical insights and practical methods to enhance fairness in active ranking systems, improving their reliability in hiring, recommendations, and other applications.
Temporal Restoration and Spatial Rewiring for Source-Free Multivariate Time Series Domain Adaptation
Source-Free Domain Adaptation (SFDA) aims to adapt a pre-trained model from an annotated source domain to an unlabelled target domain without accessing the source data, thereby preserving data privacy. While existing SFDA methods have proven effective in reducing reliance on source data, they struggle to perform well on multivariate time series (MTS) due to their failure to consider the intrinsic spatial correlations inherent in MTS data. These spatial correlations are crucial for accurately representing MTS data and preserving invariant information across domains. To address this challenge, we propose Temporal Restoration and Spatial Rewiring (TERSE), a novel and concise SFDA method tailored for MTS data. Specifically, TERSE comprises a customized spatial-temporal feature encoder designed to capture the underlying spatial-temporal characteristics, coupled with both temporal restoration and spatial rewiring tasks to reinstate latent representations of the temporally masked time series and the spatially masked correlated structures. During the target adaptation phase, the target encoder is guided to produce spatially and temporally consistent features with the source domain by leveraging the source pre-trained temporal restoration and spatial rewiring networks. Therefore, TERSE can effectively model and transfer spatial-temporal dependencies across domains, facilitating implicit feature alignment. In addition, as the first approach to simultaneously consider spatial-temporal consistency in MTS-SFDA, TERSE can also be integrated as a versatile plug-and-play module into established SFDA methods. Extensive experiments on three real-world time series datasets demonstrate the effectiveness and versatility of our approach. Our code is available at https://github.com/Tokenmw/TERSE-master.
The demand for machine learning (ML) model training on edge devices is escalating due to data privacy and personalized service needs. However, we observe that current on-device model training is hampered by the under-utilization of on-device data, due to low training throughput, limited storage and diverse data importance. To improve data resource utilization, we propose a two-stage data selection framework Titan to select the most important data batch from streaming data for model training with guaranteed efficiency and effectiveness. Specifically, in the first stage, Titan filters out a candidate dataset with potentially high importance in a coarse-grained manner. In the second stage of fine-grained selection, we propose a theoretically optimal data selection strategy to identify the data batch with the highest model performance improvement to current training round. To further enhance time-and-resource efficiency, Titan leverages a pipeline to co-execute data selection and model training, and avoids resource conflicts by exploiting idle computing resources. We evaluate Titan on real-world edge devices and three representative edge computing tasks with diverse models and data modalities. Empirical results demonstrate that Titan achieves up to 43% reduction in training time and 6.2% increase in final accuracy with minor system overhead, such as data processing delay, memory footprint and energy consumption.
With the increasing computation of training graph neural networks (GNNs) on large-scale graphs, graph condensation (GC) has emerged as a promising solution to synthesize compact, substitute graphs of the large-scale original graphs for efficient GNN training. However, these condensed graphs are specifically designed for the node classification task, significantly limiting the versatility of the synthesized data across various downstream tasks. This limitation predominantly stems from the reliance of existing GC methods on classification as the surrogate task for optimization, which leads to an excessive dependence on node labels and restricts their utility in label-scarcity scenarios. More critically, this surrogate task tends to overfit class-specific information within the condensed graph, consequently restricting the generalization capabilities of GC for other downstream tasks. To address these challenges, we introduce Contrastive Graph Condensation (CTGC), which adopts a self-supervised surrogate task to extract critical, causal information from the original graph and enhance the cross-task generalizability of the condensed graph. Specifically, CTGC employs a dual-branch framework to disentangle the generation of the node attributes and graph structures, where a dedicated structural branch is designed to explicitly encode geometric information through nodes' positional embeddings. By implementing an alternating optimization scheme with contrastive loss terms, CTGC promotes the mutual enhancement of both branches and facilitates high-quality graph generation through the model inversion technique. Extensive experiments demonstrate that CTGC excels in handling various downstream tasks with a limited number of labels, consistently outperforming state-of-the-art GC methods.
Probabilistic time series imputation has been widely applied in real-world scenarios due to its ability for uncertainty estimation and denoising diffusion probabilistic models (DDPMs) have achieved great success in probabilistic time series imputation tasks with its power to model complex distributions. However, current DDPM-based probabilistic time series imputation methodologies are confronted with two types of challenges: 1) The backbone modules of the denoising parts are not capable of achieving sequence modeling with low time complexity. 2) The architecture of denoising modules can not handle the dependencies in the time series data effectively. To address the first challenge, we explore the potential of state space model, namely Mamba, as the backbone denoising module for DDPMs. To tackle the second challenge, we carefully devise several SSM-based blocks for time series data modeling. Experimental results demonstrate that our approach can achieve state-of-the-art time series imputation results on multiple real-world datasets. Our datasets and code are available at https://github.com/decisionintelligence/SSD-TS/
Sequential recommendation (SR) is widely deployed in e-commerce platforms, streaming services, etc., revealing significant potential to enhance user experience. The core of SR lies in exploring the sequential relationships in historical user-item interactions. However, existing methods often overlook two critical factors: irregular user interests between interactions and highly uneven item distributions over time. The former factor implies that actual user preferences are not always continuous, and long-term historical interactions may not be relevant to current purchasing behavior. Therefore, relying only on these historical interactions for recommendations may result in a lack of user interest at the target time. The latter factor, characterized by peaks and valleys in interaction frequency, may result from seasonal trends, special events, or promotions. These externally driven distributions may not align with individual user interests, leading to inaccurate recommendations. To address these deficiencies, we propose TGODE to both enhance and capture the long-term historical interactions. Specifically, we first construct the user time graph and item evolution graph, which utilize user personalized preferences and global item distribution information, respectively. To tackle the temporal sparsity caused by irregular user interactions, we design a time-guided diffusion generator to automatically obtain an augmented time-aware user graph. Additionally, we devise a user interest truncation factor to efficiently identify sparse time intervals and achieve balanced preference inference. After that, the augmented user graph and item graph are fed into a generalized graph neural ordinary differential equation (ODE) to align with the evolution of user preferences and item distributions. This allows two patterns of information evolution to be matched over time. Experimental results demonstrate that TGODE outperforms baseline methods across five datasets, with improvements ranging from 10% to 46%. The code is available at https://github.com/Qin-lab-code/TGODE.
In this paper, we propose HLSAD, a novel method for detecting anomalies in time-evolving simplicial complexes. While traditional graph anomaly detection techniques have been extensively studied, they often fail to capture changes in higher-order interactions that are crucial for identifying complex structural anomalies. These higher-order interactions can arise either directly from the underlying data itself or through graph lifting techniques. Our approach leverages the spectral properties of Hodge Laplacians of simplicial complexes to effectively model multi-way interactions among data points. By incorporating higher-dimensional simplicial structures into our method, our method enhances both detection accuracy and computational efficiency. Through comprehensive experiments on both synthetic and real-world datasets, we demonstrate that our approach outperforms existing graph methods in detecting both events and change points.
Information diffusion prediction is fundamental for forecasting user participation in information sharing on social networks, such as retweets on Twitter. Existing methods typically extract user relationships from social networks and historical interactions, while further capturing contextual information within the specific diffusion process. However, these methods have several limitations: (1) They often utilize sequential diffusion process for prediction and simplify differentiated influences among participants; (2) They capture user relationships on the entire graph for all users, in which most information is not necessary for a specific diffusion process and is too inefficient for real-world large-scale networks. To tackle these limitations, we propose a novel and scalable model SILN, for sphere-based information diffusion prediction on large social networks. Specifically, SILN features three components. First, we integrate two kinds of sphere effects in terms of structural and temporal views, which learn an enhanced cascade representation. Second, SILN designs an efficient learning scheme based on the cascade-specific subgraph, which significantly reduces the entire graph computation to smaller subgraphs. Third, to facilitate subgraph extraction, we develop an optimized graph storage technique to allow constant-time neighbor access and reduce the storage cost by about 30% in practice. Extensive experiments on six real-world datasets validate that SILN consistently outperforms seven state-of-the-art competitors in prediction performance while exhibiting exceptional time and space efficiency on million-node social networks.
Accurately predicting user trading behavior in decentralized exchanges is essential for investors to mitigate risks and optimize their trading strategies. While existing research primarily focuses on predicting trading behavior in stock markets, these methods often struggle to adapt to the distinct nature of cryptocurrency trading. Specifically, they face issues such as limited adaptivity to high-frequency and algorithmic trading, as well as an insufficient consideration of fine-grained real-time market participants' behavior.Thanks to the pending mechanism of blockchain, it becomes possible to capture traders' interactions before transactions are finalized, providing valuable insights into market state. However, accurately modeling and predicting trading behavior in decentralized exchanges presents challenges, including limited adaptability to high-frequency trading, a lack of fine-grained transaction data, and high computational costs. This work proposes CryptoMixer, a lightweight fine-grained market information-aware multilayer perceptron (MLP)-based model for high-frequency cryptocurrency trading behavior prediction. Specifically, to overcome the sparsity and asynchrony of user behavior data, CryptoMixer develops a Market Information Augmenter that aggregates historical transaction data of users. Furthermore, CryptoMixer designs a Market Information Mixer as well as a Two-stream MLP Fusion Mixer to capture fine-grained user trading behavior patterns. We evaluate CryptoMixer on real-world user trading datasets from the Uniswap decentralized finance platform. Experimental results demonstrate that CryptoMixer outperforms traditional prediction models while maintaining low computational overheads, providing a practical solution for real-time cryptocurrency trading behavior prediction. The code is available at https://github.com/aqua111000/CryptoMixer.