WWW 2025 Papers

The advancement of intelligent transportation systems has led to a growing demand for accurate path representations, which are essential for tasks such as travel time estimation, path ranking, and trajectory analysis. However, traditional path representation learning (PRL) methods often focus solely on single-modal road network data, overlooking important physical and regional factors that influence real-world traffic dynamics. To overcome this limitation, we introduce Path-LLM, a multi-modal path representation learning model that integrates large language models (LLMs) into PRL. Our approach leverages LLMs to interpret both topological and textual data, enabling robust multi-modal path representations. To effectively align and merge these modalities, we propose TPalign, a contrastive learning-based pretraining strategy that ensures alignment within the embedding space. We then present TPfusion, a multimodal fusion module that dynamically adjusts the weight of each modality before integration. To further optimize LLM training, we introduce a \textit{Two-stage Overlapping Curriculum Learning (TOCL)} approach, which progressively increases the complexity of the training data. Finally, we evaluate Path-LLM on two real-world datasets across traditional PRL downstream tasks, achieving up to a 61.84\% improvement in path ranking performance on the Xi'an dataset. Additionally, Path-LLM demonstrates superior performance in both few-shot and zero-shot learning scenarios. Our code is available at: https://anonymous.4open.science/r/Path-LLM-F053.

Social recommendations (SRs) aim to enhance preference modeling by integrating social networks. However, their effectiveness is mainly constrained by two factors: the noisy social connections that may not reflect shared interests, and the limited number of social connections for most users, which hampers the system's ability to fully leverage social influence. Therefore, it is essential to perform social network refinement by removing noisy connections and adding meaningful ones for robust SRs. Inspired by the denoising capability of generative diffusion models, we propose a Model-Agnostic Social Network Refinement framework with Diffusion Models for Robust Social Recommendation (ARD-SR). Specifically, in the forward process, we corrupt the social network by progressively adding position-specific Gaussian noise calibrated to the user preference similarity, better simulating how the social network responds to noise perturbations. The reverse process learns to denoise, guided by each user’s neighborhood preferences from the SR backbone, generating a tailored social network aligned with each user's preference for establishing connections. For effective learning, we design a curriculum-based training mechanism that progressively introduces challenging samples characterized by high sparsity or high noise levels. Finally, ARD-SR and the SR backbone are alternately trained, ensuring a continuous mutual enhancement between the social network refinement and the backbone's user representation learning. To further enhance the quality of the refined social network, (1) we introduce a preference-guided flip operation during inference to improve the input quality; and (2) we modify social connections based on the exponential weighted moving average of ARD-SR's predictions across epochs to reduce fluctuations. Experiments on three datasets show that ARD-SR significantly improves SR performance across multiple SR backbones.

The foundation model has heralded a new era in artificial intelligence, pretraining a single model to offer cross-domain transferability on different datasets. Graphs are omnipresent non-Euclidean structures, ranging from recommender systems to biochemical structures. Graph neural networks excel at learning graph data, but often lack the generalization capacity. Hence, graph foundation model is drawing increasing attention, and recent efforts have been made to leverage Large Language Models, encouraged by the remarkable success of GPT-4. On the one hand, existing studies primarily focus on text-attributed graphs, while a wider range of real graphs do not contain fruitful textual attributes. On the other hand, the sequential graph description tailored for the Large Language Model neglects the structural complexity, which is a predominant characteristic of the graph. Such limitations motivate an important question: Can we go beyond Large Language Model, and pretrain a universal model to learn the structural knowledge for any graph? The answer in the language or vision domain is a shared vocabulary. We observe the fact that there also exist shared substructures underlying the graph domain, and thereby open the new opportunity of graph foundation model with structure vocabulary (by which any graph can be constructed). The key innovation of this paper is the discovery of a simple yet effective structural vocabulary of trees and cycles, and we explore its inherent connection to Riemannian geometry. Herein, we present a universal pretraining model, RiemannGFM, with geometric contrastive learning. Concretely, we first construct a novel product bundle to incorporate the diverse geometries of the vocabulary. On this constructed space, we stack Riemannian layers where the structural vocabulary, regardless of specific graph, is learnt in Riemannian manifold. This offers the shared structural knowledge for cross-domain transferability, and node encoding is generated in the tangent bundle for arbitrary input graph. Empirical results show the superiority of RiemannGFM on a diversity of real graphs.

The online advertising market has recently reached the 500 billion dollar mark, and to accommodate the need to match a user with the highest bidder at a fraction of a second, it has moved towards a complex, automated and often opaque model that involves numerous agents and intermediaries. Stimulated by the lack of transparency, but also the enormous potential profits, bad actors have found ways to circumvent restrictions, and generate substantial revenue that can support websites with objectionable or even illegal content. In this work, we evaluate transparency Web standards and shed light on how shady actors take advantage of gaps to absorb ad revenues while putting the brand safety of advertisers in danger. We collect and study a large corpus of over 7 million websites and show how ad transparency standards can be abused by bad actors to obscure ad revenue flows. We show how identifier pooling can redirect ad revenues from reputable domains to notorious domains serving objectionable content and that the phenomenon is underestimated by previous studies by a factor of 15. Finally, we publish a Web monitoring service that enhances the transparency of supply chains and business relationships among Web entities.

Large Language Models (LLMs) are increasingly employed in zero-shot documents ranking, yielding commendable results. However, several significant challenges still persist in LLMs for ranking: (1) LLMs are constrained by limited input length, precluding them from processing a large number of documents simultaneously; (2) The output document sequence is influenced by the input order of documents, resulting in inconsistent ranking outcomes; (3) Achieving a balance between cost and ranking performance is quite challenging. To tackle these issues, we introduce a novel documents ranking method called TourRank, which is inspired by the tournament mechanism. This approach alleviates the impact of LLM's limited input length through intelligent grouping, while the tournament-like points system ensures robust ranking, mitigating the influence of the document input sequence. We test TourRank with different LLMs on the TREC DL datasets and the BEIR benchmark. Experimental results show that TourRank achieves state-of-the-art performance at a reasonable cost.

This paper addresses the online $k$-selection problem with diseconomies of scale (OSDoS), where a seller seeks to maximize social welfare by optimally pricing items for sequentially arriving buyers, accounting for increasing marginal production costs. Previous studies have investigated deterministic dynamic pricing mechanisms for such settings. However, significant challenges remain, particularly in achieving optimality with small or finite inventories and developing effective randomized posted price mechanisms. To bridge this gap, we propose a novel randomized dynamic pricing mechanism for OSDoS, providing a tighter lower bound on the competitive ratio compared to prior work. Our approach ensures optimal performance in small inventory settings (i.e., when $k$ is small) and surpasses existing online mechanisms in large inventory settings (i.e., when $k$ is large), leading to the best-known posted price mechanism for optimizing online selection and allocation with diseconomies of scale across varying inventory sizes.

Communities in networks are groups of nodes that are more densely connected to each other than to the rest of the network, forming clusters with strong internal relationships. When nodes have sensitive attributes, such as demographic groups in social networks, a key question is whether nodes in each group are equally well-connected within each community. We model connectivity fairness through group modularity, an adaptation of modularity that accounts for group structures. We introduce two versions of group modularity grounded on different null models and present fairness-aware community detection algorithms. Finally, we provide experimental results on real and synthetic networks, evaluating both the group modularity of community structure in networks and our fairness-aware algorithms.

Explainable recommendation has demonstrated significant advantages in informing users about the logic behind recommendations, thereby increasing system transparency, effectiveness, and trustworthiness. To provide personalized and interpretable explanations, existing works often combine the generation capabilities of large language models (LLMs) with collaborative filtering (CF) information. CF information extracted from the user-item interaction graph captures the user behaviors and preferences, which is crucial for providing informative explanations. However, due to the complexity of graph structure, effectively extracting the CF information from graphs still remains a challenge. Moreover, existing methods often struggle with the integration of extracted CF information with LLMs due to its implicit representation and the modality gap between graph structures and natural language explanations. To address these challenges, we propose G-Refer, a framework using Graph Retrieval-augmented large language models (LLMs) for explainable recommendation. Specifically, we first employ a hybrid graph retrieval mechanism to retrieve explicit CF signals from both structural and semantic perspectives. The retrieved CF information is explicitly formulated as human-understandable text by the proposed graph translation and accounts for the explanations generated by LLMs. To bridge the modality gap, we introduce knowledge pruning and retrieval-augmented fine-tuning to enhance the ability of LLMs to process and utilize the retrieved CF information to generate explanations. Extensive experiments show that G-Refer achieves superior performance compared with existing methods in both explainability and stability. Codes and data are available at https://anonymous.4open.science/r/G-Refer.

The Semantic Web and Graph Database communities have developed three distinct schema languages for RDF and graph-structured data: SHACL, ShEx, and PG-Schema. Each language has its unique approach to defining constraints and validating graph data. In this work, we provide formal, concise definitions of the core components of each of these schema languages. We employ a uniform framework to facilitate a comprehensive comparison between the languages and identify a common set of functionalities, shedding light on both overlapping and distinctive features of the three languages.

Semi-supervised anomaly detection (AD) has garnered growing attention due to its ability to effectively combine limited labeled data with abundant unlabeled data. However, current methods of-ten impose artificial constraints on the proportion of unlabeled anomalies in the training set or overlook potential noise from these anomalies, thereby impeding the effective training of models for anomaly detection in real-world scenarios where several anomalies may be present in the unlabeled dataset. Additionally, existing methods often struggle to effectively exploit and model the complex relationships between data instances, which is critical for learning more discriminative features and accurate distance measures. Distance-based methods, in particular, typically rely on Euclidean distance metric, which lacks the flexibility to capture complex correlations across different data dimensions. To address above challenges, we propose CAD, a denoising-aware Contrastive distance learning framework for semi-supervised AD. It introduces a contrastive training objective to facilitate the learning of distinctive representations by contrasting the average distance between anomalies and unlabeled samples. To fully exploit the information from the unlabeled data meanwhile mitigate the effects of noise, we incorporate a two-stage anomaly denoising and expansion strategy to refine the dataset by identifying high-confidence samples from the unlabeled set. Furthermore, we employ a parameterized bilinear tensor distance layer to learn a customized distance metric, enabling the model to capture intricate relationships among data points. Extensive experiments on 10 real-world datasets demonstrate that CAD significantly outperforms existing semi-supervised AD models. Code available at https://github.com/CADrepo/CAD.

The web experience in developing regions remains subpar, primarily due to the growing complexity of modern webpages and insufficient optimization by content providers. Users in these regions typically rely on low-end devices and limited bandwidth, which results in a poor user experience as they download and parse webpages bloated with excessive third-party CSS and JavaScript (JS). To address these challenges, we introduce the Mobile Application Markup Language (MAML), a flat layout-based web specification language that reduces computational and data transmission demands, while replacing the excessive bloat from JS with a new scripting language centered on essential (and popular) web functionalities. Last but not least, MAML is backward compatible as it can be transpiled to minimal HTML/JavaScript/CSS and thus work with legacy browsers. We benchmark MAML in terms of page load times and sizes, using a translator which can automatically port any webpage to MAML. When compared to the popular Google AMP, across 100 testing webpages, MAML offers webpage speedups by tens of seconds under challenging network conditions thanks to its significant size reductions. Next, we run a competition involving 25 university students porting 50 of the above webpages to MAML using a web-based editor we developed. This experiment shows that, with little developer effort, MAML is quite effective in maintaining the visual and functional correctness of the originating webpages.

The remarkable ability of diffusion models to generate high-fidelity images has led to their widespread adoption. However, concerns have also arisen regarding their potential to produce Not Safe for Work (NSFW) content and exhibit social biases, impeding their practical use and progress in real-world applications. In response to this challenge, prior work has primarily focused on employing security filters to identify and subsequently exclude toxic text, or alternatively, fine-tuning pre-trained diffusion models to erase sensitive concepts. Unfortunately, existing methods struggle to achieve satisfactory performance in the sense that they can have a significant impact on the normal model output while still failing to prevent the generation of harmful content in some cases. In this paper, we propose a novel self-discovery approach to identifying a semantic direction vector in the embedding space to restrict text embedding within a safe region. Our method circumvents the need for correcting individual words within the input text and steers the entire text prompt towards a safe region in the embedding space, thereby enhancing model robustness against all possibly unsafe prompts. In addition, we employ a Low-Rank Adaptation (LoRA) for semantic direction vector initialization to reduce the impact on the model performance for other semantics. Furthermore, our method can also be integrated with existing methods to improve their socially responsible performance. Extensive experiments on benchmark datasets demonstrate that our method can effectively reduce NSFW content and mitigate social bias generated by diffusion models compared to several state-of-the-art baselines.

Delivering superior search services is crucial for enhancing cus- tomer experience and driving revenue growth in e-commerce. Con- ventionally, search systems model user behaviors by combining user preference and query-item relevance statically, often through a fixed logical ‘and’ relationship. This paper reexamines existing approaches through a unified lens using both causal graphs and Venn diagrams, uncovering two prevalent yet significant issues: entangled preference and relevance effects, and a collapsed model- ing space. To surmount these challenges, our research introduces a novel framework, DRP, which enhances search accuracy through two components to reconstruct the behavior modeling space. Specif- ically, we implement preference editing to proactively remove the relevance effect from preference predictions, yielding untainted user preferences. Additionally, we employ adaptive fusion, which dynamically adjusts fusion criteria to align with the varying pat- terns of relevance and preference, facilitating more nuanced and tailored behavior predictions within the reconstructed modeling space. Empirical validation on two public datasets and a propri- etary e-commerce search dataset underscores the superiority of our proposed methodology, demonstrating marked improvements in performance over existing approaches.

Multi-label text classification (MLTC) aims to assign multiple relevant labels to a given sentence. An inherent challenge of MLTC is capturing label correlations compared with multi-class text classification. Existing MLTC models primarily focus on leveraging correlation information but often overlook the common issue of overfitting. Meanwhile, plug-and-play regularization methods struggle to preserve correlations effectively. In this paper, we distinguish two types of label correlations: explicit co-occurring correlation and implicit semantic correlations, and propose two regularization methods based on prototypical label embeddings for two correlation preservation, respectively. Specifically, we first generate the prototypical label embedding of multiple co-occurred labels as an intermediate. We then apply a prototypical label regularization on the distance between the sentence embedding and corresponding prototypical label embedding to alleviate the over-alignment issue caused by binary cross entropy loss and facilitate explicit correlation preservation. We finally extend the vanilla Mixup, which solely mixes multi-hot labels, on prototypical label embedding mixing to promote implicit correlation preservation. Empirical studies show the effectiveness of our regularization methods.

While internet usage is slowly catching up globally, it is still unclear how the web experience differs in developing and developed countries. On the one hand, the web has a notoriously large inertia, with many websites still relying on unencrypted HTTP, deprecated web features, or old and buggy libraries. On the other hand, developing countries are expected to leapfrog and directly adopt the newest technologies by learning from the prior mistakes of more developed countries. Anecdotal evidence suggests that websites in developing and developed regions differ significantly. In this work, we test this hypothesis by measuring differences in web development practices across the two groups of countries, using multiple dimensions: websites' size, complexity, security, privacy, quality, technology adoption, and accessibility. Concretely, we collect the largest dataset to date that compares web development practices across developed and developing regions -- 200,000 websites across 20 countries -- which we aim to open source along with this publication. Our findings reveal that websites in developing regions are generally smaller and simpler, utilizing fewer requests — an adaptation that improves the performance over slower network conditions common in these areas. However, these sites are less optimized in other critical aspects: they frequently employ inefficient image formats, include unnecessary JavaScript code, lack responsive image designs, and offer limited accessibility features for individuals with disabilities. Notably, our security assessment shows developing regions lagging in HTTPS adoption and vulnerability mitigation, possibly due to lower awareness of best practices.

Large Language Models (LLMs) have achieved impressive results in various tasks but struggle with hallucination problems and lack of relevant knowledge, especially in deep complex reasoning and knowledge-intensive tasks. Knowledge Graphs (KGs), which capture vast amounts of facts in a structured format, offer a reliable source of knowledge for reasoning. However, existing KG-based LLM reasoning methods face challenges like handling multi-hop reasoning, multi-entity questions, and effectively utilizing graph structures. To address these issues, we propose Paths-over-Graph (PoG), a novel method that enhances LLM reasoning by integrating knowledge reasoning paths from KGs, improving the interpretability and faithfulness of LLM outputs. PoG tackles multi-hop and multi-entity questions through a three-phase dynamic multi-hop path exploration, which combines the inherent knowledge of LLMs with factual knowledge from KGs. In order to improve the efficiency, PoG prunes irrelevant information from the graph exploration first and introduces efficient three-step pruning techniques that incorporate graph structures, LLM prompting, and a pre-trained language model (e.g., SBERT) to effectively narrow down the explored candidate paths. This ensures all reasoning paths contain highly relevant information captured from KGs, making the reasoning faithful and interpretable in problem-solving. PoG innovatively utilizes graph structure to prune the irrelevant noise and represents the first method to implement multi-entity deep path detection on KGs for LLM reasoning tasks. Comprehensive experiments on five benchmark KGQA datasets demonstrate PoG outperforms the state-of-the-art method ToG across GPT-3.5-Turbo and GPT-4, achieving an average accuracy improvement of 18.9\%. Notably, PoG with GPT-3.5-Turbo surpasses ToG with GPT-4 by up to 23.9\%.

Blockchain interoperability protocols enable cross-chain asset transfers or data retrievals between isolated chains, which are considered as the core infrastructure for Web 3.0 applications such as decentralized finance protocols. However, existing protocols either face severe scalability issues due to high on-chain and off-chain costs, or suffer from trust concerns because of centralized designs. In this paper, we propose \texttt{MAP}, a trustless blockchain interoperability protocol that relays cross-chain transactions across heterogeneous chains with high scalability. First, within \texttt{MAP}, we develop a novel cross-chain relay technique, which integrates a unified relay chain architecture and on-chain light clients of different source chains, allowing the retrieval and verification of diverse cross-chain transactions. Furthermore, we reduce cross-chain verification costs by incorporating an optimized zk-based light client scheme that adaptively decouples signature verification overheads from inefficient smart contract execution and offloads them to off-chain provers. For experiments, we conducted the first large-scale evaluation on existing interoperability protocols. With \texttt{MAP}, the required number of on-chain light clients is reduced from $O(N^2)$ to $O(N)$, with around 35\% reduction in on-chain costs and 25\% reduction for off-chain costs when verifying cross-chain transactions. To demonstrate the effectiveness, we deployed \texttt{MAP} in the real world. By 2024, we have supported over six popular public chains, 50 cross-chain applications and relayed over 200K cross-chain transactions worth over 640 million USD. Based on rich practical experiences, we constructed the first real-world cross-chain dataset to further advance blockchain interoperability research.

Clustering over a graph seeks to partition the nodes therein into disjoint groups such that nodes within the same cluster are tightly-knit, while those across clusters are distant from each other. In practice, graphs are often attended with rich attributes, which are termed attributed graphs. By leveraging the complementary nature of graph topology and node attributes in such graphs, graph neural networks (GNNs) have obtained encouraging performance in graph clustering. However, existing GNN-based approaches strongly rely on the homophilic assumption of the input graph, and thus, largely fail on heterophilic graphs and others embodying numerous missing or noisy links, which are widely present in real life. To bridge this gap, this paper presents DGAC, an effective graph-agnostic solution for graph clustering. Particularly, DGAC overcomes the limitations of prior works by exploiting the high-order connectivity of nodes within not only the input graph G but also the affinity graph H underlying the attribute data. To achieve this goal, we first unify the embedding and clustering generations into a coherent framework that optimizes Dirichlet Energy on both G and H. Based thereon, theoretical-grounded solvers are developed for efficient constructions of the embeddings and clusters, which capture high-order semantics from G or H via graph diffusion. On top of that, DGAC includes three optimization loss functions that facilitate effective feature extraction and clustering. Extensive experiments, comparing DGAC against 12 baselines over 12 homophilic or heterophilic graph datasets, showcase that DGAC consistently and considerably outperforms all competitors in terms of clustering quality measured against ground truth labels.

Network connectivity minimization is a fundamental problem in controlling the spread of epidemics and facilitating information propagation in social networks. The problem aims to identify a budget number of key nodes whose removal would minimize the connectivity of a network. However, the existing solutions heavily rely on the number of edges, making it challenging to handle large and densely connected social networks. In this study, we present a fast algorithm that is independent of the number of edges. To achieve this, we first introduce a surrogate matrix that approximates the residual adjacency matrix with arbitrary small predefined error. We then devise an efficient approach for calculating the key nodes by optimizing the eigenvalues of the surrogate matrix. Remarkably, the algorithm has a small time complexity , with a small tunable number. Our algorithm thereby maintains a linear scalability in terms of the number of nodes and is unaffected by the number of edges. Hence, it has the capability to efficiently handle large and dense social networks. At last, we evaluate its performance against state-of-the-art techniques using diverse real-world datasets. The experimental results demonstrate the superiority of our proposed method in terms of both solution quality and computational efficiency.