WWW 2025 Papers

Privacy-preserving Cross-Domain Recommendation (CDR) has been extensively studied to address the cold-start problem using auxiliary source domains while simultaneously protecting sensitive information. However, existing privacy-preserving CDR methods rely heavily on transferring sensitive user embeddings or behaviour logs, which leads to adopt privacy methods to distort the data patterns before transferring it to the target domain. The distorted information can compromise overall performance during the knowledge transfer process. To overcome these challenges, our approach differs from existing privacy-preserving methods that focus on safeguarding user-sensitive information. Instead, we concentrate on distilling transferable knowledge from insensitive item embeddings, which we refer to as \textbf{prototypes}. Specifically, we propose a conditional model inversion mechanism to accurately distill prototypes for individual users. We have designed a new data format and corresponding learning paradigm for distilling transferable prototypes from traditional recommendation models using model inversion. These prototypes facilitate bridging the domain shift between distinct source and target domains in a privacy-friendly manner. Additionally, they enable the identification of top-k users in the target domain to substitute for cold-start users prediction. We conduct extensive experiments across large real-world datasets, and the results substantiate the effectiveness of PFCDR.

Federated Graph Learning (FedGL) is an emerging Federated Learning (FL) framework that learns the graph data from various clients to train better Graph Neural Networks(GNNs) model. Owing to concerns regarding the security of such framework, numerous studies have attempted to execute backdoor attacks on FedGL, with a particular focus on distributed backdoor attacks. However, all existing methods posting distributed backdoor attack on FedGL only focus on injecting distributed backdoor triggers into the training data of each malicious client, which will cause model performance degradation on original task and is not always effective when confronted with robust federated learning defense algorithms, leading to low success rate of attack. What's more, the backdoor signals introduced by the malicious clients may be smoothed out by other clean signals from the honest clients, which potentially undermining the performance of the attack. To address the above significant shortcomings, we propose a non-intrusive graph distributed backdoor attack(NI-GDBA) that does not require backdoor triggers to be injected in the training data. Our attack trains an adaptive perturbation trigger generator model for each malicious client to learn the natural backdoor from the GNN model downloading from the server with the malicious client's local data. In contrast to traditional distributed backdoor attacks on FedGL via trigger injection in training data, our attack on different datasets such as Molecules and Bioinformatics have higher attack success rate, stronger persistence and stealth, and has no negative impact on the performance of the global GNN model. We also explore the robustness of NI-GDBA under different defense strategies, and based on our extensive experimental studies, we show that our attack method is robust to current federated learning defense methods, thus it is necessary to consider non-intrusive distributed backdoor attacks on FedGL as a novel threat that requires custom defenses. Code is available at an anonymous github repository: https://anonymous.4open.science/r/NI-GDBA-64E5/

Deceptive patterns (DPs) are user interface designs deliberately crafted to manipulate users into unintended decisions, often by exploiting cognitive biases for the benefit of companies or services. While numerous studies have explored ways to identify these deceptive patterns, many existing solutions require significant human intervention and struggle to keep pace with the evolving nature of deceptive designs. To address these challenges, we expanded the deceptive pattern taxonomy from security and privacy perspectives, refining its categories and scope. We created a comprehensive dataset of deceptive patterns by integrating existing small-scale datasets with new samples, resulting in 6,725 images and 10,421 DP instances from mobile apps and websites. We then developed DPGuard, a novel automatic tool leveraging commercial multimodal large language models (MLLMs) for deceptive pattern detection. Experimental results show that DPGuard outperforms state-of-the-art methods. Finally, we conducted an extensive empirical evaluation on 2,000 popular mobile apps and websites, revealing that 23.61% of mobile screenshots and 47.27% of website screenshots feature at least one deceptive pattern instance. Through four unexplored case studies that inform security implications, we highlight the critical importance of the unified taxonomy in addressing the growing challenges of Internet deception.

Graph anomaly detection plays a crucial role in identifying nodes that deviate significantly from normal patterns within a graph, with applications spanning various domains such as fraud detection, authorship fraud, and rumor propagation. Traditional methods primarily focus on aggregating information from neighboring nodes and reconstructing the central node based on these aggregated features. The anomaly degree is then calculated by comparing the reconstructed features with the original ones. Despite their effectiveness, these methods face limitations due to the constraints of device performance and the need to protect user privacy. In reality, graph data is often partitioned and distributed across different local clients, which leads to isolated client subgraphs. This partitioning results in incomplete feature aggregation, as the connections between subgraphs are missing, ultimately reducing the performance of anomaly detection models. To overcome these challenges, a federated graph anomaly detection approach based on disentangled representation learning is proposed. This method separates node features into two distinct components: intrinsic features and subgraph style features. By identifying outliers within the subgraph style features, a set of pseudo-nodes is generated and shared across the entire graph. These pseudo-nodes simulate connections between otherwise isolated subgraphs, which enables more comprehensive aggregation of intrinsic features from neighboring nodes. In addition, conditional variational autoencoders (CVAE) are employed alongside contrastive learning strategies to alleviate class imbalance and achieve effective feature disentanglement. These techniques help ensure that anomalous nodes are detected more accurately despite the inherent challenges of federated graph systems. Extensive experiments conducted on six diverse datasets provide compelling evidence of the proposed method's superior performance in federated graph anomaly detection, highlighting its ability to effectively handle incomplete graph structures while maintaining data privacy.

Phishing attacks pose a significant threat to Internet users. Understanding the security posture of phishing infrastructure is crucial for developing effective defense strategies, as it helps identify potential weaknesses that attackers might exploit. Despite extensive research, there may still be a gap in fully understanding these security weaknesses. To address this important issue, this paper presents a longitudinal study of security configurations and vulnerabilities in phishing websites and associated kits. We focus on two main areas: (1) analyzing the security configurations of phishing websites and servers, particularly HTTP headers and application-level security, and (2) examining the prevalence and types of vulnerabilities in phishing kits. We analyze data from 906,731 distinct phishing websites collected over 2.5 years, covering HTML headers, client-side resources, and phishing kits. Our findings suggest that phishing websites often employ weak security configurations, with 88.8% of the 13,344 collected phishing kits containing at least one potential vulnerability, and 12.5% containing backdoor vulnerabilities. These vulnerabilities present an opportunity for defenders to shift from passive defense to active disruption of phishing operations. Our research proposes a new approach to leverage weaknesses in phishing infrastructure, allowing defenders to take proactive actions to disable phishing sites earlier and reduce their effectiveness.

In the digital age, resources such as open-source software and publicly accessible databases form a crucial category of digital public goods, providing extensive benefits across the Internet. However, the inherent non-exclusivity and non-competitiveness of these public goods frequently result in under-provision, a dilemma exacerbated by individuals' tendency to free-ride. This scenario fosters both cooperation and competition among users, leading to the emergence of public goods games. This paper investigates networked public goods games involving heterogeneous players and convex costs to explore solutions of Nash Equilibrium (NE) for this problem. In these games, each player can choose her own effort level, representing the contributions to public goods. We employ network structures to depict the interactions among participants. Each player's utility is composed of a \emph{concave} value component, influenced by collective efforts, and a \emph{convex} cost component, determined solely by individual effort. To the best of our knowledge, this study is the first to explore a networked public goods game with convex costs. Our research begins by examining welfare solutions aimed at maximizing social welfare and ensuring the convergence of pseudo-gradient ascent dynamics. We establish the presence of NE in this model and provide an in-depth analysis of the conditions under which NE is unique. Additionally, we introduce the concept of game equivalence, which expands the range of public goods games that can support a unique NE. We also delve into \emph{comparative statics}, an essential tool in economics, to evaluate how slight modifications in the model—interpreted as monetary redistribution—impact player utilities. In addition, we analyze a particular scenario with a predefined game structure, illustrating the practical relevance of our theoretical insights. Consequently, our research enhances the broader understanding of strategic interactions and structural dynamics in networked public goods games, with significant implications for policy design in internet economic and social networks.

OWL (Web Ontology Language) ontologies, which are able to represent both relational and type facts as standard knowledge graphs and complex domain knowledge in Description Logic (DL) axioms, are widely adopted in domains such as healthcare and bioinformatics. Inspired by the success of knowledge graph embeddings, embedding OWL ontologies has gained significant attention in recent years. Current methods primarily focus on learning embeddings for atomic concepts and roles, enabling the evaluation based on normalized axioms through specially designed score functions. However, they often neglect the embedding of complex concepts, making it difficult to infer with more intricate axioms. This limitation reduces their effectiveness in advanced reasoning tasks, such as Ontology Learning and ontology-mediated Query Answering. In this paper, we propose $\mathcal{EL}^{++}$-closed ontology embeddings which are able to represent any logical expressions in DL $\mathcal{EL}^{++}$ via composition. Furthermore, we develop TransBox, an effective $\mathcal{EL}^{++}$-closed ontology embedding method that can handle many-to-one, one-to-many and many-to-many relations. Our extensive experiments demonstrate that TransBox often achieves state-of-the-art performance across various real-world datasets for predicting complex axioms. Code and data are available at the https://anonymous.4open.science/r/TransBox-F4B7.

Terms and conditions for online shopping websites often contain terms that can have significant financial consequences for customers. Despite their impact, there is currently no comprehensive understanding of the types and potential risks associated with unfavorable financial terms. Furthermore, there are no publicly available detection systems or datasets to systematically identify or mitigate these terms. In this paper, we take the first steps toward solving this problem with three key contributions. First, we introduce TermMiner, an automated data collection and topic modeling pipeline to understand the landscape of unfavorable financial terms. Second, we create ShopTC-100K, a dataset of terms and conditions from shopping websites in the Tranco top 100K list, comprising 1.8 million terms from 8,251 websites. Consequently, we develop a taxonomy of 22 types from 4 categories of unfavorable financial terms—spanning purchase, post-purchase, account termination, and legal aspects. Third, we build TermLens, an automated detector that uses Large Language Models (LLMs) to identify unfavorable financial terms. Fine-tuned on an annotated dataset, TermLens achieves an F1 score of 94.6% and a false positive rate of 2.3% using GPT-4o. When applied to shopping websites from the Tranco top 100K, we find that 47.21% of these sites contain at least one unfavorable financial term, with such terms being more prevalent on less popular websites. Case studies further highlight the financial risks and customer dissatisfaction associated with unfavorable financial terms, as well as the limitations of existing ecosystem defenses.

Graph Neural Networks (GNNs) are vulnerable to backdoor attacks, where adversaries implant malicious triggers to manipulate model predictions. Existing graph backdoor attacks are susceptible to defense mechanisms or robust classifiers because they rely on subgraph injection or structural perturbations, e.g., creating additional edges to attach backdoor triggers to the original graph. To enhance the stealthiness of graph backdoors, we propose SPEAR, a novel structure-preserving graph backdoor attack that avoids modifying the graph’s topology. SPEAR operates within a limited attack budget by selectively perturbing node attributes while ensuring the triggers exert significant influence through a global importance-driven feature selection strategy. Additionally, a neighborhood-aware trigger generator is employed to underpin a high attack success rate by utilizing semantic information from the neighborhood. SPEAR amplifies effectiveness and stealthiness by combining subtle yet impactful attribute manipulation with a refined trigger generation mechanism. Extensive experiments demonstrate that SPEAR achieves state-of-the-art effectiveness in bypassing defenses on real-world datasets, establishing it as a potent and stealthy backdoor attack for graph-based tasks.

Bayer-patterned color filter array (CFA) has been the go-to solution for color image sensors. In augmented reality (AR), although color interpolation (i.e. demosaicing) of pre-demosaic RAW images facilitates user-friendly rendering, it creates no benefits in offloaded neural network analytics but only increases the image channels by $3\times$ with higher transmission overheads. Thus, we propose ABO, an adaptive RAW frame offloading framework that parallelizes demosaicing with DNN offloading. The contributions are three-fold: First, we design a configurable tile-wise RAW image neural codec to compress frame sizes while sustaining the downstream DNN accuracy under various bandwidth restraints. Second, based on content-aware tiles-in-frame selection and runtime bandwidth estimation, a dynamic transmission controller adaptively calibrates codec configurations to maximize the DNN accuracy under real-time constraints. Third, we further optimize the system pipelining to reduce the end-to-end frame processing latency. Through extensive evaluations on a prototype platform, ABO consistently provides a 40\% more frame processing throughput and a 30\% less end-to-end latency while improving the offloaded DNN accuracy by up to 15\% compared to SOTA baselines. It also presents improved robustness against dim light and motion blur situations.

Facility location problems on graphs are ubiquitous in the real world and hold significant importance, yet their resolution is often impeded by NP-hardness. MIP solvers can find the optimal solutions but fail to handle large instances, while algorithm efficiency has a higher priority in cases of emergency. Recently, machine learning methods have been proposed to tackle such classical problems with fast inference, but they are limited to the myopic constructive pattern and only consider simple cases in Euclidean space. This paper introduces a unified and generalizable approach to tackle facility location problems on weighted graphs with deep reinforcement learning, demonstrating a keen awareness of complex graph structures. Striking a harmonious balance between solution quality and running time, our method stands out with superior efficiency and steady performance. Our model trained on small graphs is highly scalable and consistently generates high-quality solutions, achieving a speedup of more than 2000 times to Gurobi on instances with 1000 nodes. The experiments on Shanghai road networks further demonstrate its practical value in solving real-world problems.

In response to stringent data protection regulations, websites typically display a cookie banner to inform users about the usage and purposes of cookies, seeking their explicit consent before installing any cookies into their browsers. However, a systematic approach for reliably assessing compliance between the website-declared purpose and the semantic-intended purpose of cookies (denoted as $potential$ $cookie$ $purpose$ $violation$) has been notably absent. Websites may still, whether intentionally or unintentionally (e.g., due to third-party libraries imported), mis-declare cookies that may be abused for tracking purposes. We address this gap with COOVER ($\underline{coo}kie$ $\underline{v}alue$ $examin\underline{er}$). We advocate that the value of the cookie is a more reliable indicator of its semantic-intended purpose compared to other features, such as expires and meta-information, which can be easily obfuscated. COOVER decomposes the cookie value into primitive $segments$ representing minimal semantic units, and fine-tunes a GPT-3.5 model to automatically interpret their semantics. Based on the interpretation, it classifies cookies into four GDPR-defined purposes. We benchmark COOVER against two widely-used content management providers (CMPs) i.e., CookiePedia and Cookie Script, and the state-of-the-art cookie classifier named CookieBlock. It achieves an F1 score of 95%, significantly outperforming other methods. To understand the $status$ $quo$ of potential cookie purpose violations on the web, we employ COOVER to analyze Alexa Top 1k websites. Remarkably, out of 15,339 cookies across these websites, only 3.1% quality as $truly$ necessary cookies, while 44.1% of websites suffer from issues of potential purpose violation. Our work serves as a wake-up call to web service providers and encourages further regulatory interventions to rectify non-compliance issues within the web infrastructure.

Large language models (LLM) achieve superior performance in generative tasks. However, due to the natural gap between language model generation and structured information extraction in three dimensions: task type, output format, and modeling granularity, they often fall short in structured information extraction, a crucial capability for effective data utilization on the web. In this paper, we define the generation process of the language model as the controllable state transition, aligning the generation and extraction processes to ensure the integrity of the output structure and adapt to the goals of the information extraction task. Furthermore, we propose the Structure2Text decider to help the language model understand the fine-grained extraction information, which converts the structured output into natural language and makes state decisions, thereby focusing on the task-specific information kernels, and alleviating language model hallucinations and incorrect content generation. We conduct extensive experiments and detailed analyses on myriad information extraction tasks. Our method not only achieves significant performance improvements but also ensures the integrity of the output structure, making it easy to parse the extracted content.

Dynamics modeling has been introduced as a novel paradigm in message passing (MP) of graph neural networks (GNNs). Existing methods consider MP between nodes as a heat diffusion process, and leverage \textit{heat equation} to model the temporal evolution of nodes in the embedding space. However, heat equation can hardly depict the wave nature of graph signals in graph signal processing. Besides, heat equation is essentially a partial differential equation (PDE) involving a first partial derivative of time, whose numerical solution usually has low stability, and leads to inefficient model training. In this paper, we would like to depict more wave details in MP, since graph signals are essentially wave signals that can be seen as a superposition of a series of waves in the form of eigenvector. This motivates us to consider MP as a wave propagation process to capture the temporal evolution of wave signals in the space. Based on wave equation in physics, we innovatively develop a graph wave equation to leverage the wave propagation on graphs. In details, we demonstrate that the graph wave equation can be connected to traditional spectral GNNs, facilitating the design of graph wave networks (GWNs) based on various Laplacians and enhancing the performance of the spectral GNNs. Besides, the graph wave equation is particularly a PDE involving a second partial derivative of time, which has stronger stability on graphs than the heat equation that involves a first partial derivative of time. Additionally, we theoretically prove that the numerical solution derived from the graph wave equation are constantly stable, enabling to significantly enhance model efficiency while ensuring its performance. Extensive experiments show that GWNs achieve state-of-the-art and efficient performance on benchmark datasets, and exhibit outstanding performance in addressing challenging graph problems, such as over-smoothing and heterophily. Our code is available at https://anonymous.4open.science/r/GWN/.

Decentralized learning (DL) is an emerging technique that allows nodes on the web to collaboratively train machine learning models without sharing raw data. Dealing with stragglers, i.e., nodes with slower compute or communication than others, is a key challenge in DL. We present DivShare, a novel asynchronous DL algorithm that achieves fast model convergence in the presence of communication stragglers. DivShare achieves this by having nodes fragment their models into parameter subsets and send, in parallel to computation, each subset to a random sample of other nodes instead of sequentially exchanging full models. The transfer of smaller fragments allows more efficient usage of the collective bandwidth and enables nodes with slow network links to contribute with at least some of their model parameters quickly. By theoretically proving the convergence of DivShare, we provide, to the best of our knowledge, the first formal proof of convergence for a DL algorithm that accounts for the effects of asynchronous communication with delays. We experimentally evaluate DivShare against two state-of-the-art DL baselines, AD-PSGD and Swift, and with two standard datasets, CIFAR-10 and Movielens. We find that DivShare with communication stragglers lowers time-to-accuracy by up to 3.9x compared to AD-PSGD on the CIFAR-10 dataset. Compared to baselines, DivShare also achieves up to 19.4% better accuracy and 9.5% lower test loss on the CIFAR-10 and Movielens datasets, respectively.

With the adoption of deep learning models to low-power, small-memory edge devices, energy consumption and storage usage of such models has become a key concern. The problem acerbates even further with ever-growing data and equally-matched bulkier models. This concern is particularly pronounced for graph data due to its quadratic storage, irregular (non-grid) geometry, and very large size. Typical graph data, such as road networks, infrastructure networks, social networks easily exceeds millions of nodes, and several gigabytes of storage is needed just to store the node embedding vectors, let alone the model parameters. In recent years, the memory issue has been addressed by moving away from memory-intensive double precision floating-point arithmetic towards single-precision or even half-precision, often by trading-off marginally small performance. Along this effort, we propose Node2binary, which embeds graph nodes in as low as 128 binary bits, which drastically reduces the memory footprint of vertex embedding vectors by several order of magnitude. Node2binary leverages a fast community detection algorithm to covert the given graph into a hierarchical partition tree and then find embedding of graph vertices in binary space by solving a combinatorial optimization (CO) task over the tree edges. CO is NP-hard, but Node2binary uses an innovative combination of discrete gradient descent and randomization to solve this effectively and efficiently. Our extensive experiments over four real-world graphs show that Node2binary achieves competitive performances compared to the state-of-the art graph embedding methods in both node classification and link prediction tasks.

Search result diversification (SRD), aimed at ensuring that selected documents in a ranking list cover a wide range of subtopics, is a significant and extensively studied problem in Web search and Information Retrieval. Existing methods primarily utilize a paradigm of "greedy selection", i.e., selecting one document with the highest diversity score at a time. These approaches tend to be inefficient and are easily trapped in a suboptimal state. In addition, some other methods optimize an approximation of the objective function, but the results still remain suboptimal. To address these challenges, we introduce \textbf{M}ulti-\textbf{A}gent reinforcement learning (MARL) for search result \textbf{DIV}ersity, which called \textbf{MA4DIV}. In this approach, each document is an agent and the search result diversification is modeled as a cooperative task among multiple agents. By modeling the SRD ranking problem as a cooperative MARL problem, this approach allows for directly optimizing the diversity metrics, such as $\alpha$-NDCG, while achieving high training efficiency. We conducted experiments on public TREC datasets and a large-scale dataset in the industrial setting. The results show that MA4DIV achieves substantial improvements in both effectiveness and efficiency than existing baselines, especially on the industrial scale dataset.

Advertisers are increasingly using automated bidding to optimize their ad campaigns on online advertising platforms. Autobidding allows an advertiser to optimize her objective subject to various constraints. In this paper, we design online autobidding algorithms to optimize value subject to ROI and budget constraints. We consider an item is being auctioned in each of $T$ rounds. We focus on one buyer with budget and ROI constraints in the stochastic setting: her value and highest competing bid faced are drawn i.i.d. from some unknown (joint) distribution in each round. We design low-regret bidding algorithms that bid on behalf of this buyer. Our main result is an algorithm with full information feedback (i.e., the highest competing bid is revealed after each round) that guarantees a near-optimal $\tilde O(\sqrt T)$ regret with respect to the best Lipschitz function that maps values to bids. The class of Lipschitz bidding functions is rich enough to best respond to many correlation structures between value and highest competing bid, e.g., positive or negative correlation. Our result applies to a wide range of auctions, most notably any mixture of first- and second-price auctions. In addition, our result holds for both value-maximizing buyers and quasi-linear utility-maximizing buyers. We also study the bandit setting, where the algorithm only observes whether the bidder wins the auction or not. In this setting, we show an $\Omega(T^{2/3})$ regret lower bound for first-price auctions, showing a significant disparity between the full information and bandit settings. We also design an algorithm with a regret bound of $\tilde O(T^{3/4})$ when the value distribution is known and is independent of the highest competing bid.

In today's web-scale, data-driven environments, real-time detection of persistent items that consistently recur over time is essential for maintaining system integrity, reliability, and security. Persistent items often signal critical anomalies, such as stealthy DDoS and botnet attacks in web infrastructures. Although various methods exist for identifying such items as well as for determining their frequency, they require recording every item for processing, which is impractical at very high data rates achieved by modern data streams. In this paper, we introduce Pontus, a novel approach that uses an approximate data structure (sketch) specifically designed for the efficient and accurate detection of persistent items. Our method not only achieves fast and precise lookup but is also flexible, allowing for minor modifications to accommodate other types of persistence-based item detection tasks, such as detecting persistent items with low frequency. We rigorously validate our approach through formal methods, offering detailed proofs of time/space complexity and error bounds to demonstrate its theoretical soundness. Our extensive trace-driven evaluations across various persistence-based tasks further demonstrate Pontus's effectiveness in significantly improving detection accuracy and enhancing processing speed compared to existing approaches. We implement Pontus in an experimental platform with industry-grade Intel Tofino switches and demonstrate the practical feasibility of our approach in a real-world memory-constrained environment.