WWW 2025 论文列表

We propose a novel framework that addresses a critical limitation in Graph Self-Supervised Learning (GSSL) for graph classification: the underestimation of topological information. Traditional GSSL, despite its success in various benchmarks, often fails to fully leverage the expressive power of Graph Neural Networks (GNNs), particularly in capturing complex structural properties. This limitation stems from two main factors: (1) the inadequacy of conventional GNNs in representing sophisticated topological features, and (2) the focus of self-supervised learning solely on final graph representations. To address these issues, we introduce GenHopNet, a GNN framework that integrates a k-hop message-passing scheme, enhancing its ability to capture local structural information without explicit substructure extraction. We theoretically demonstrate that GenHopNet surpasses the expressiveness of the classical Weisfeiler-Lehman (WL) test for graph isomorphism. Furthermore, we propose a structural- and positional-aware GSSL framework that incorporates topological information throughout the learning process. This approach enables the learning of representations that are both sensitive to graph topology and invariant to specific structural and feature augmentations. Comprehensive experiments on graph classification datasets, including those designed to test structural sensitivity, show that our methods consistently outperform most of the existing approaches in accuracy while maintaining computational efficiency. Our work significantly advances GSSL's capability in distinguishing graphs with similar local structures but different global topologies.

Stress haunts people in modern society, which may cause severe health issues if left unattended. With social media becoming an integral part of daily life, leveraging social media to detect stress has gained increasing attention. While the majority of the work focuses on classifying stress states and stress categories, this study introduce a new task aimed at estimating more specific stressors (like exam, writing paper, etc.) through users' posts on social media. Unfortunately, the diversity of stressors with many different classes but a few examples per class, combined with the consistent arising of new stressors over time, hinders the machine understanding of stressors. To this end, we cast the stressor estimation problem within a practical scenario few-shot learning setting, and propose a novel meta-learning based stressor estimation framework that is enhanced by a meta-knowledge inheritance mechanism. This model can not only learn generic stressor context through meta-learning, but also has a good generalization ability to estimate new stressors with little labeled data. A fundamental breakthrough in our approach lies in the inclusion of the meta-knowledge inheritance mechanism, which equips our model with the ability to prevent catastrophic forgetting when adapting to new stressors. The experimental results show that our model achieves state-of-the-art performance compared with the baselines. Additionally, we construct a social media-based stressor estimation dataset that can help train web mining models to facilitate human well-being.

With the development of social media, people are exposed to a vast amount of unverified information, making fact-checking particularly important. Existing fact-checking methods primarily encourage breaking down claims into more easily solvable sub-tasks, and deriving final answers through reasoning with external evidence. However, these models face logical issues regarding whether and how the sub-tasks can logically be combined to form the original claims, and encounter causal errors in the reasoning process due to insufficient evidence or hallucinations from LLMs. In addition, they often suffer from a lack of interpretability. In this paper, we propose $\textbf{Lo}$gical and $\textbf{Ca}$usa$\textbf{l}$ fact-checking (LoCal), a novel fact-checking framework based on multiple LLM-based agents. The usage of multi-agent systems is due to their increasingly demonstrated ability to perform complex tasks in a manner similar to humans. LoCal primarily consists of a decomposing agent, multiple reasoning agents, and two evaluating agents. Specifically, the decomposing agent first utilizes the in-context learning ability of LLMs to break down complex claims into simpler sub-tasks, including fact verification tasks and question answering tasks. Afterwards, two types of reasoning agents are respectively utilized to retrieve external knowledge to address the fact verification tasks that requires comparative analysis skills, and the question answering tasks that necessitates the ability of information extraction from evidence. We then combine the sub-tasks and their corresponding responses to generate a solution for evaluation. In order to enhance logical and causal consistency, two evaluating agents are respectively employed to examine whether the generated solution is logically equivalent to the original claim and determine whether the solution still hold when challenged by the counterfactual label. The evaluating agents provide confidence degrees for the solutions based on the evaluation results and iteratively correct the logical and causal errors in the reasoning process. We evaluate LoCal on two challenging datasets, and the results show that LoCal significantly outperforms all the baseline models across different settings of evidence availability. In addition, LoCal offers better interpretability by providing a structured solution along with detailed evaluating processes. We believe LoCal will provide valuable insights for future agent-based misinformation detection.

Knowledge Graph Query Embedding (KGQE) aims to embed First-Order Logic (FOL) queries in a low-dimensional KG space for complex reasoning over incomplete KGs. To enhance the generalization of KGQE models, recent studies integrate various external information (such as entity types and relation context) to better capture the logical semantics of FOL queries. The whole process is commonly referred to as Query Pattern Learning (QPL). However, current QPL methods typically suffer from the pattern-entity alignment bias problem, leading to the learned defective query patterns limiting KGQE models' performance. To address this problem, we propose an effective Query Instruction Parsing Plugin (QIPP) that leverages the context awareness of Pre-trained Language Models (PLMs) to capture latent query patterns from code-like query instructions. Unlike the external information introduced by previous QPL methods, we first propose code-like instructions to express FOL queries in an alternative format. This format utilizes textual variables and nested tuples to convey the logical semantics within FOL queries, serving as raw materials for a PLM-based instruction encoder to obtain complete query patterns. Building on this, we design a query-guided instruction decoder to adapt query patterns to KGQE models. To further enhance QIPP's effectiveness across various KGQE models, we propose a query pattern injection mechanism based on compressed optimization boundaries and an adaptive normalization component, allowing KGQE models to utilize query patterns more efficiently. Extensive experiments demonstrate that our plug-and-play method improves the performance of eight basic KGQE models and outperforms two state-of-the-art QPL methods.

Quantitative information about entities constitutes a significant portion of tabular data in open sources and data lakes. Tuch tables often lack consistent labeling and proper schema, posing significant challenges for querying and integration. This paper studies the problem of numerical column annotation in scenarios where quantitative data may be gathered from different sources and unit consistency is a concern. For instance, weight measurements may vary between entities, expressed in kilograms for some and pounds for others, with no accompanying unit information. We investigate the conditions for effectively annotating mixed-unit numeric data, introduce a benchmark for such an annotation task, and propose an algorithm that reliably detects semantic types (e.g., height) and links them to the corresponding types present in a knowledge graph. Our evaluation on a diverse set of columns with mixed units and varying levels of annotation difficulty shows that our method significantly outperforms strong baselines such as GPT-4o-mini and SAND in terms of accuracy, excelling in both detecting mixed units and annotating them with appropriate semantic labels. All our code and data will be publicly released upon acceptance of the paper.

The classic clique model has properties of hereditaries and cohesiveness. Here hereditaries means a subgraph of a clique is still a clique. Counting small cliques in a graph is a fundamental operation of numerous applications. However, the clique model is often too restrictive for practical use, leading to the focus on other relaxed-cliques with properties of hereditaries and cohesiveness. To address this issue, we investigate a new problem of counting general hereditary cohesive subgraphs (\hcs). All subgraphs with properties of hereditaries and cohesiveness can be called a kind of \hcs. To count \hcs, we propose a general framework called \hcspivot, which can be applied to count all kinds of \hcs. \hcspivot can count most \hcs combinatorially without explicitly listing them. Two additional noteworthy features of \hcspivot are its ability to (1) simultaneously count \hcs of any size and (2) simultaneously count \hcs for each node or each edge. Based on our \hcspivot framework, we propose two novel algorithms with several carefully designed pruning techniques to count $s$-defective cliques and $s$-plexes, which are two specific types of \hcs. We conduct extensive experiments on 8 large real-world graphs, and the results demonstrate the high efficiency and effectiveness of our solutions.

In text-based person search endeavors, data generation has emerged as a prevailing practice, addressing concerns over privacy preservation and the arduous task of manual annotation. Although the number of synthesized data can be infinite in theory, the scientific conundrum persists that how much generated data optimally fuels subsequent model training. We observe that only a subset of the data in these constructed datasets plays a decisive role. Therefore, we introduce a new Filtering-WoRA paradigm, which contains a filtering algorithm to identify this crucial data subset and WoRA (Weighted Low-Rank Adaptation) learning strategy for light fine-tuning. The filtering algorithm is based on the cross-modality relevance to remove the lots of coarse matching synthesis pairs. As the number of data decreases, we do not need to fine-tune the entire model. Therefore, we propose a WoRA learning strategy to efficiently update a minimal portion of model parameters. WoRA streamlines the learning process, enabling heightened efficiency in extracting knowledge from fewer, yet potent, data instances. Extensive experimentation validates the efficacy of pretraining, where our model achieves advanced and efficient retrieval performance on challenging real-world benchmarks. Notably, on the CUHK-PEDES dataset, we have achieved a competitive mAP of 67.02% while reducing model training time by 19.82%.

Recent advances in recommender systems have shown that user-system interaction essentially formulates long-term optimization problems, and online reinforcement learning can be adopted to improve recommendation performance. The general solution framework incorporates a value function that estimates the user's expected cumulative rewards in the future and guides the training of the recommendation policy. To avoid local maxima, the policy may explore potential high-quality actions during inference to increase the chance of finding better future rewards. To accommodate the stepwise recommendation process, one widely adopted approach to learning the value function is learning from the difference between the values of two consecutive states of a user. However, we argue that this paradigm involves an incorrect approximation in the stochastic process. Specifically, between the current state and the next state in each training sample, there exist two separate random factors from the stochastic policy and the uncertain user environment. Original TD learning under these mixed random factors may result in a suboptimal estimation of the long-term rewards. As a solution, we show that these two factors can be separately approximated by decomposing the original temporal difference loss. The disentangled learning framework can achieve a more accurate estimation with faster learning and improved robustness against action exploration. As empirical verification of our proposed method, we conduct offline experiments with online simulated environments built based on public datasets.

Graphs are able to model interconnected entities in many online services, supporting a wide range of applications on the Web. This raises an important question: How can we train a graph foundational model on multiple source domains and adapt to an unseen target domain? A major obstacle is that graphs from different domains often exhibit divergent characteristics. Some studies leverage large language models to align multiple domains based on textual descriptions associated with the graphs, limiting their applicability to text-attributed graphs. For text-free graphs, very few recent works attempt to align different feature distributions across domains, while generally neglecting structural differences. In this work, we propose a novel Structure Alignment framework for text-free Multi-domain Graph Pre-Training and cross-domain adaptation (SAMGPT). It is designed to learn multi-domain knowledge from graphs originating in multiple source domains, which can then be adapted to address applications in an unseen target domain. Specifically, we introduce a set of structure tokens to harmonize structure-based aggregation across source domains during the pre-training phase. Next, for cross-domain adaptation, we design dual prompts, namely, holistic prompts and specific prompts, which adapt unified multi-domain structural knowledge and fine-grained, domain-specific information, respectively, to a target domain. Finally, we conduct comprehensive experiments on seven public datasets to evaluate and analyze the effectiveness of SAMGPT. (Codes and data are available at https://anonymous.4open.science/r/SAMGPT for anonymous review.)

Large language models (LLMs), with advanced linguistic capabilities, have been employed in reranking tasks through a sequence-to-sequence approach. In this paradigm, multiple passages are reranked in a listwise manner and a textual reranked permutation is generated. However, due to the limited context window of LLMs, this reranking paradigm requires a sliding window strategy to iteratively handle larger candidate sets. This not only increases computational costs but also restricts the LLM from fully capturing all the comparison information for all candidates. To address these challenges, we propose a novel self-calibrated listwise reranking method, which aims to leverage LLMs to produce global relevance scores for ranking. To achieve it, we first propose the relevance-aware listwise reranking framework, which incorporates explicit list-view relevance scores to improve reranking efficiency and enable global comparison across the entire candidate set. Second, to ensure the comparability of the computed scores, we propose self-calibrated training that uses point-view relevance assessments generated internally by the LLM itself to calibrate the list-view relevance assessments. Extensive experiments and comprehensive analysis on the BEIR benchmark and TREC Deep Learning Tracks demonstrate the effectiveness and efficiency of our proposed method.

Graph condensation has emerged as an intriguing technique to save the expensive training costs of Graph Neural Networks (GNNs) by substituting a condensed small graph with the original graph. Despite the promising results achieved, previous methods usually employ an entangled paradigm of redundant parameters (nodes, edges, GNNs), which incurs complex joint optimization during condensation. This paradigm has considerably impeded the scalability of graph condensation, making it challenging to condense extremely large-scale graphs and generate high-fidelity condensed graphs. Therefore, we propose to disentangle the condensation process into a two-stage GNN-free paradigm, independently condensing nodes and generating edges while eliminating the need to optimize GNNs at the same time. The node condensation module avoids the complexity of GNNs by focusing on node feature alignment with anchors of the original graph, while the edge translation module constructs the edges of the condensed nodes by transferring the original structure knowledge with neighborhood anchors. This simple yet effective approach achieves at least 10 times faster than state-of-the-art methods with comparable accuracy on medium-scale graphs. Moreover, the proposed DisCo can successfully scale up to the Ogbn-papers100M graph containing over 100 million nodes with flexible reduction rates and improves performance on the second-largest Ogbn-products dataset by over 5\%. Extensive downstream tasks and ablation study on five common datasets further demonstrate the effectiveness of the proposed DisCo framework. The source code will be made publicly available.

Retrieval plays a fundamental role in recommendation systems, search, and natural language processing (NLP) by efficiently finding relevant items from a large corpus given a query. Dot products have been widely used as the similarity function in such retrieval tasks, thanks to Maximum Inner Product Search (MIPS) that enabled efficient retrieval based on dot products. However, state-of-the-art retrieval algorithms have migrated to learned similarities. Such algorithms vary in form; the queries can be represented with multiple embeddings, complex neural networks can be deployed, the item ids can be decoded directly from queries using beam search, and multiple approaches can be combined in hybrid solutions. Unfortunately, we lack efficient solutions for retrieval in these state-of-the-art setups. Our work investigates techniques for efficient retrieval with expressive learned similarity functions. We first prove that Mixture-of-Logits (MoL) is a universal approximator, and can express all learned similarity functions. We then demonstrate how to apply MoL to common retrieval tasks in recommendation systems and NLP. We next propose techniques to retrieve the approximate top $K$ results using MoL with a tight bound. We finally compare our techniques with existing approaches, showing that MoL, with a new mutual information-based load balancing loss we propose, sets new state-of-the-art results across heterogeneous scenarios, including sequential retrieval models in recommendation systems and finetuning language models for question answering; and our approximate top-k retrieval with learned similarities outperforms baselines by up to 105× in latency, while achieving > .99 recall rate of exact algorithms.

Influence maximization (IM) aims to identify a small set of influential users to maximize the information spread. It has been widely applied in the context of viral marketing, where a company distributes incentives to a few influencers to promote the product. However, in practical scenarios, not all users hold equal importance and certain users need to be prioritized for the specific requirements. Motivated by this, recently, a variant problem of IM, called influence maximization with priority (IMP), has been proposed. Given a graph G=(V,E), a priority set P ⊆ V and a threshold T ∈ [0, |P|], IMP aims to identify a set of k nodes (termed seeds) to maximize the expected number of activated nodes in G while satisfying that the expected number of activated nodes in P is no less than the given threshold. Nevertheless, we show that existing solutions for IMP are inferior in maximizing the influence spread in G, and can only offer poor approximation ratios in many cases. To address these limitations, in this paper, we first propose a novel framework named SAR with both superior empirical effectiveness and strong theoretical guarantees. In addition, to obtain more practical results, we study the IMP problem under the adaptive setting, where the seed users are iteratively selected after observing the diffusion result of the previous seeds. We design a scalable and effective algorithm AAS that achieves expected approximation guarantees. Comprehensive experiments on 5 real-world datasets are conducted to validate the performance of the proposed techniques. Compared with the state-of-the-art method, SAR achieves up to 22.3% larger spread and AAS achieves up to 42.6% larger spread, with both exhibiting a higher empirical approximation ratio.

WebGL has long been the prevalent API for GPU-accelerated graphics in web browsers, boosting 2D/3D graphical web applications. Despite widespread adoption, WebGL's programming model hinders its rendering performance on modern GPU hardware. To this end, WebGPU has been proposed as the next-generation API of GPU-accelerated processing in web browsers, exhibiting higher performance than WebGL. However, considering the complex logic of WebGL applications and the still-evolving WebGPU specification, statically migrating existing WebGL applications to WebGPU from source code is labor-intensive. To address this issue, we propose GL2GPU, an intermediate layer that dynamically translates WebGL to WebGPU at JavaScript runtime to improve rendering performance. GL2GPU addresses the inconsistencies between the WebGL and WebGPU programming models by emulating WebGL rendering states and leverages performance optimization mechanisms introduced by WebGPU to reduce the overhead of dynamic translation. Evaluation of three representative WebGL benchmarks shows that GL2GPU significantly enhances end-to-end rendering performance while maintaining visual consistency, achieving an average frame time reduction of 45.05\% across different devices and operating systems.

WebAssembly (Wasm) is a universal low-level bytecode designed to build modern web systems. Recent studies have shown that technologies such as voltage scaling and RowHammer attacks are expected to increase the likelihood of bit flips, which may cause unacceptable or catastrophic system failures. This raises concerns about the impact of bit flips on Wasm programs, which run as instructions in web systems, and it is an undeveloped topic since the features of Wasm differ from traditional programs. In this paper, we propose a novel paradigm, namely IVPSEG, to understand the error propagation of bit flips within Wasm programs. Specifically, we first use Large Language Models (LLMs) to automatically extract instruction embeddings containing semantic knowledge of each instruction's context. Then, we exploit these embeddings and program structure (control execution and data transfer) to construct a semantic enhanced code property graph, which implicates the potential path of error propagation. Based on this graph, we utilize graph neural networks and attention diffusion to optimize instruction embeddings by capturing different error propagation patterns for instruction vulnerability prediction. In particular, we build a Wasm compilation and fault generation system to simulate bit flips at Wasm runtime. Our experimental results with 14 benchmark programs and test cases show IVPSEG outperforms the state-of-the-art methods in terms of accuracy (average 13.06\%$\uparrow$ ), F1-score (average 14.93\%$\uparrow$), and model robustness.

Recommender systems are vital for personalizing user experiences, yet they often rely on implicit feedback data that can be noisy and misleading. Existing denoising studies typically involve either incorporating auxiliary information or learning denoising strategies from interaction data. Nonetheless, they face challenges due to the inherent limitations of external knowledge and interaction data, as well as the non-universality of certain predefined assumptions, which hinder their ability to accurately identify noise. Recently, large language models (LLMs) have garnered significant attention due to their extensive world knowledge and powerful reasoning capabilities. Despite this, the potential of LLMs to enhance the denoising process in recommendations remains largely unexplored. In this paper, we introduce LLaRD, a novel framework that leverages LLMs to improve the denoising process in recommender systems, thereby enhancing overall recommendation performance. Specifically, LLaRD generates denoising-related knowledge by first enriching semantic insights from observational data through LLMs, facilitating a comprehensive inference of user-item preference knowledge. It then employs a novel Chain-of-Thought (CoT) technique over user-item interaction graphs to uncover relation knowledge pertinent to denoising. Finally, it utilizes the Information Bottleneck (IB) principle to align the denoising knowledge generated by LLMs with the recommendation targets, effectively filtering out both data noise and irrelevant knowledge produced by the LLMs. Empirical results demonstrate the effectiveness of our proposed framework, showcasing its superior performance in denoising and recommendation accuracy.

In applications such as data management and Web search engines, indexes are key to enabling efficient data retrieval. We find that unlike standard benchmarks with uniform data distribution, index operations in real-world tasks often exhibit strong skewness. However, existing high-performance learned indexes, while proposed to enhance query and update efficiency, often fail to account for the characteristics of skewed workload access, leading to an imbalanced focus on optimizing a single performance metric at the expense of other critical aspects of overall index performance. Furthermore, the complete use of learned models in index structures can lead to increased robustness issues, making them highly vulnerable to attacks and resulting in system unavailability. To address these challenges, we propose ShapeShifter, an adaptive evolutionary structure based on traditional indexes, capable of dynamically adjusting node structures according to the workload. ShapeShifter introduces a node evolution strategy, designed with workload-skew-aware policies, to adaptively adjust and optimize the most suitable partial index structure, leveraging a hybrid mechanism that combines traditional and learned structures for robust performance and an optimal time-space trade-off under skewed workloads and extreme data conditions. The evaluation results show that ShapeShifter achieves the optimal trade-off between performance and space efficiency while maintaining robustness.

Damage analysis in social media platforms such as Twitter is a comprehensive problem which involves different subtasks for mining damage-related information from tweets e.g., informativeness, humanitarian categories and severity assessment). The comprehensive information obtained by damage analysis enables to identify breaking events around the world in real-time and hence provides aids in emergency responses. Recently, with the rapid development of web technologies, multimodal damage analysis has received increasing attentions due to users' preference of posting multimodal information in social media. Multimodal damage analysis leverages the associated image modality to improve the identification of damage-related information in social media. However, existing works on multimodal damage analysis address each damage-related subtask individually and do not consider their joint training mechanism. In this work, we propose the Bidirectional Multi-task Cascaded multimodal Fusion (BiMCF) approach towards joint multimodal damage analysis. To this end, we introduce the cascaded multimodal fusion framework to separately integrate effective visual and text information for each task, considering that different tasks attend to different information. To exploit the interactions across tasks, bidirectional propagation of the attended image-text interactive information is implemented between tasks, which can lead to enhanced multimodal fusion. Comprehensive experiments are conducted to validate the effectiveness of the proposed approach.

Recent studies have demonstrated the effectiveness of using large language language models (LLMs) in passage ranking. The listwise approaches, such as RankGPT, have become new state-of-the-art in this task. However, the efficiency of RankGPT models is limited by the maximum context length and relatively high latency of LLM inference. To address these issues, in this paper, we propose PE-Rank, leveraging the single passage embedding as a good context compression for efficient listwise passage reranking. By treating each passage as a special token, we can directly input passage embeddings into LLMs, thereby reducing input length. Additionally, we introduce an inference method that dynamically constrains the decoding space to these special tokens, accelerating the decoding process. For adapting the model to reranking, we employ listwise learning to rank loss for training. Evaluation results on multiple benchmarks demonstrate that PE-Rank significantly improves efficiency in both prefilling and decoding, while maintaining competitive ranking effectiveness.