With the growing scale of social media, social event detection and evolution modeling have attracted increasing attention. Graph neural networks (GNNs) and transformer-based pre-trained language models (PLMs) have become mainstream approaches in this area. However, existing methods still face three major challenges. First, the sheer volume of social media messages makes learning resource-intensive. Second, the fragmentation of social media messages often impedes the model's ability to capture a comprehensive view of the events. Third, the lack of structured temporal context has hindered the development of effective models for event evolution, limiting users' access to event information. To address these challenges, we propose a foundation model for unsupervised Social Event Detection and Evolution, namely RagSEDE. Specifically, RagSEDE introduces a representativeness- and diversity-driven sampling strategy to extract key messages from massive social streams, significantly reducing noise and computational overhead. It further establishes a novel paradigm based on Retrieval Augmented Generation (RAG) that enhances PLMs in detecting events while simultaneously constructing and maintaining an evolving event knowledge base. Finally, RagSEDE leverages structural information theory to dynamically model event evolution keywords for the first time. Extensive experiments on two public datasets demonstrate the superiority of RagSEDE in open-world social event detection and evolution.
论文检索
输入标题、作者或关键词，从 5,485 篇学术成果中精准定位
Multimodal knowledge graphs (MMKG) extend the representational capacity of conventional knowledge graphs by integrating multimodal contexts. However, despite increasing attention on core tasks—multimodal link prediction (MMLP), multimodal entity alignment (MMEA), and multimodal entity linking (MMEL)—most existing approaches remain confined to meticulously tailored, task-specific and graph-specific paradigms, limiting their practicality in dynamic environments where tasks co-exist and new knowledge graphs continually emerge. In this work, we propose the first unified multi-task inductive inference framework for \MMKG (MtaIMKG) that can handle multiple core tasks, with universality and transferability to any unseen MMKG, by tackling inherent heterogeneities: i) Feature Heterogeneity: Bridging the semantic gaps in feature representations across domains. ii) Structural Heterogeneity: Overcoming structural divergence in topology across graphs. iii) Task Heterogeneity: Reconciling conflicting objectives and inductive biases across tasks. To address the above challenges, we center on reconciling divergent representation preferences across tasks, delivering only task-pertinent information instead of engaging in indiscriminate aggregation. Specifically, we introduce a task-aware routing network that dynamically integrates and routes multimodal representations, assesses their contributions under task semantics, and produces adaptive, task-conditioned representations for any input graph. In addition, MtaIMKG employs a gated mixture-of-experts to suppress cross-modal noise and extract complementary information across tasks. These strategies tackle the notable heterogeneity challenges in multi-task inductive inference on MMKG, enabling MtaIMKG to effectively zero-shot generalize to any unseen graph. Extensive experiments demonstrate that MtaIMKG achieves SOTA performance in both multi-task inductive and transductive inference with competitive efficiency, confirming its value as a scalable and generalizable solution for MMKG reasoning. Our code is released at https://github.com/MMKGer/MtaIMKG/.
Metaphors are a fundamental cognitive tool for articulating abstract and subjective experiences and implicit semantics, making them potent indicators of psychological state, particularly in individuals with depression. The proliferation of social media has created a vast repository of such metaphorical expressions, offering an unprecedented opportunity to understand mental health struggles. These metaphors can provide crucial insights for clinical assessment and therapeutic intervention. However, their potential remains largely untapped in automated depression detection, primarily due to the lack of large-scale, annotated datasets. To bridge this gap, we introduce the Depression-Related Metaphor Dataset (DRMD), a novel resource of social media posts related to depression, incorporating depression levels (severe, moderate, minimum, and null), the presence or absence of metaphors, and their conceptual source domain mappings. We leverage this dataset to fine-tune Large Language Models (LLMs), integrating metaphorical features to enhance detection capabilities. Our results demonstrate that models incorporating metaphorical information achieve superior accuracy in depression detection and, importantly, generate high-quality explanations for their decisions by referencing specific metaphorical expressions. This work underscores the critical role of metaphorical analysis in computational mental health and provides a foundation for future research in explainable AI for depression detection. The dataset is publicly available.
Copy trading has become the dominant entry strategy in meme coin markets. However, due to the market's extremely illiquid and volatile nature, the strategy exposes an exploitable attack surface: adversaries deploy manipulative bots to front-run trades, conceal positions, and fabricate sentiment, systematically extracting value from naïve copiers at scale. Despite its prevalence, bot-driven manipulation remains largely unexplored, and no robust defensive framework exists. We propose a manipulation-resistant copy-trading system based on a multi-agent architecture powered by a multi-modal large language model (LLM) and chain-of-thought (CoT) reasoning. Our approach outperforms zero-shot and most statistic-driven baselines in prediction accuracy as well as all baselines in economic performance, achieving an average copier return of 3% per meme coin investment under realistic market frictions. Overall, our results demonstrate the effectiveness of agent-based defenses and predictability of trader profitability in adversarial meme coin markets, providing a practical foundation for robust copy trading.
Network protocols are the foundation of modern communication, yet their implementations often contain semantic vulnerabilities stemming from inadequate understanding of specification semantics. Existing gray-box and black-box testing approaches lack semantic modeling of protocols, making it difficult to precisely express testing intent and cover boundary conditions. Moreover, they typically rely on coarse-grained oracles such as crashes, which are inadequate for identifying deep semantic vulnerabilities. To address these limitations, we present a semantics-aware fuzzing framework, SemFuzz. The framework leverages large language models to extract structured semantic rules from RFC documents and generates test cases that intentionally violate these rules to encode specific testing intents. It then detects deep semantic vulnerabilities by comparing the observed responses with the expected ones. Evaluation on seven widely deployed protocol implementations shows that SemFuzz identified sixteen potential vulnerabilities, ten of which have been confirmed. Among the confirmed vulnerabilities, five were previously unknown and four have been assigned CVEs. These results demonstrate the effectiveness of SemFuzz in detecting semantic vulnerabilities.
The security of web services increasingly relies on accurate detection of advanced, previously unseen attacks hidden within complex host activities. Provenance-based intrusion detection systems (PIDSes) offer a promising foundation for this task by capturing rich causal and structural relationships across processes, files, and network interactions. However, recent studies show that these graph-driven methods remain vulnerable to graph manipulation attacks, where adversaries subtly alter provenance graphs to evade detection, which limits their practical deployment. To address this challenge, we present ProvGuard, a robust anomaly detection framework that couples logic-aware multi-view augmentation with contrastive representation learning. Instead of applying arbitrary structural perturbations, ProvGuard employs Logic-Aware Noise Injection (LNI) to generate semantically valid graph views that preserve the causal semantics of provenance data. These views are then leveraged in a Logic-Preserving Contrastive Learning module, enabling the model to learn representations invariant to benign transformations yet sensitive to adversarial inconsistencies. Extensive evaluations on multiple provenance datasets show that ProvGuard surpasses state-of-the-art detectors in resisting graph manipulation attacks while maintaining high detection accuracy and efficiency, achieving an average F1-score above 96% with less than a 10% AUC drop.
Retrieval-Augmented Generation (RAG) effectively mitigates hallucinations and knowledge gaps in Large Language Models (LLMs) for knowledge-intensive tasks by incorporating external web-based knowledge. However, when integrating diverse yet potentially conflicting web-sourced information, RAG systems are prone to knowledge conflicts that manifest as incorrect or inconsistent model behaviors, ultimately leading to unreliable responses. To address this challenge, this paper proposes Conflict-Aware RAG, a general training framework that leverages the model's inherent conflict-sensing capability to build a more robust RAG system via phased optimization. At the core of this framework lies ConScore, a conflict signal that quantifies the model's awareness of potential knowledge conflicts by comparing generative probabilities across distinct knowledge sources. This signal then guides both the construction of training data and a multi-stage optimization workflow: In the Supervised Fine-Tuning (SFT) stage, conflict features are employed to select representative distracting documents, laying the groundwork for core RAG capabilities; in the Direct Preference Optimization (DPO) stage, high-quality preference pairs are constructed using the conflict signal to boost the model's robustness against distracting knowledge; and in the Reranking stage, conflict confidence and information gain are integrated to synergistically optimize the collaboration mechanism between the retriever and LLM. Experiments on six knowledge-intensive question answering (QA) datasets demonstrate that Conflict-Aware RAG significantly outperforms mainstream baselines. Further ablation studies and quantitative analyses validate the method's stability and generalization, laying the foundation for robust RAG systems.
The predictive probability of the next token (P_token) in large language models (LLMs) is inextricably linked to the probability of relevance for the next piece of information, the purchase probability of the next product, and the execution probability of the next action-all of which fall under the scope of the task-level target distribution (P_task). While LLMs are known to generate samples that approximate real-world distributions, whether their fine-grained sampling probabilities faithfully align with task requirements remains an open question. Through controlled distribution-sampling simulations, we uncover a striking dichotomy in LLM behavior, distinguishing two model types: D-models (e.g. Qwen-2.5), whose P_token exhibits large step-to-step variability and poor alignment with P_task; and E-models (e.g. Mistral-Small), whose P_token is more stable and better aligned with P_task. We further evaluate these two model types in downstream tasks such as code generation and recommendation, revealing systematic trade-offs between diversity and stability that shape task outcomes. Finally, we analyze the internal properties of both model families to probe their underlying mechanisms. These findings offer foundational insights into the probabilistic sampling behavior of LLMs and provide practical guidance on when to favor D- versus E-models. For web-scale applications, including recommendation, search, and conversational agents, our results inform model selection and configuration to balance diversity with reliability under real-world uncertainty, providing a better level of interpretation.
Previous studies on visual customization primarily rely on the objective alignment between various control signals (e.g., language, layout and canny) and the edited images, which largely ignore the subjective emotional contents, and more importantly lack general-purpose foundation models for affective visual customization. With this in mind, this paper proposes an LLM-centric Affective Visual Customization (L-AVC) task, which focuses on generating images within modifying their subjective emotions via Multimodal LLM. Further, this paper contends that how to make the model efficiently align emotion conversion in semantic (named inter-emotion semantic conversion) and how to precisely retain emotion-agnostic contents (named exter-emotion semantic retaining) are rather important and challenging in this L-AVC task. To this end, this paper proposes an Efficient and Precise Emotion Manipulating (EPEM) approach for editing subjective emotions in images. Specifically, an Efficient Inter-emotion Converting (EIC) module is tailored to make the LLM efficiently align emotion conversion in semantics before and after editing, followed by a Precise Extra-emotion Retaining (PER) module to precisely retain the emotion-agnostic contents. Comprehensive experimental evaluations on our constructed L-AVC dataset demonstrate the great advantage of the proposed EPEM approach to the L-AVC task over several state-of-the-art baselines. This justifies the importance of emotion information for L-AVC and the effectiveness of EPEM in efficiently and precisely manipulating such information.
Skin-toned emojis are crucial for fostering personal identity and social inclusion in online communication. As AI models, particularly Large Language Models (LLMs), increasingly mediate interactions on web platforms, the risk that these systems perpetuate societal biases through their representation of such symbols is a significant concern. This paper presents the first large-scale comparative study of bias in skin-toned emoji representations across two distinct model classes. We systematically evaluate dedicated emoji embedding models (emoji2vec, emoji-sw2v) against four modern LLMs (Llama, Gemma, Qwen, and Mistral). Our analysis first reveals a critical performance gap: while LLMs demonstrate robust support for skin tone modifiers, widely-used specialized emoji models exhibit severe deficiencies. More importantly, a multi-faceted investigation into semantic consistency, representational similarity, sentiment polarity, and core biases uncovers systemic disparities. We find evidence of skewed sentiment and inconsistent meanings associated with emojis across different skin tones, highlighting latent biases within these foundational models. Our findings underscore the urgent need for developers and platforms to audit and mitigate these representational harms, ensuring that AI's role on the web promotes genuine equity rather than reinforcing societal biases.
Graph Neural Networks (GNNs) have achieved remarkable success across various graph-based tasks but remain highly sensitive to distribution shifts. In this work, we focus on a prevalent yet underexplored phenomenon in graph generalization, Minimal Shift Flip (MSF)—where test samples that slightly deviate from the training distribution are abruptly misclassified. To interpret this phenomenon, we revisit MSF through the lens of Sharpness-Aware Minimization (SAM), which characterizes the local stability and sharpness of the loss landscape while providing a theoretical foundation for modeling generalization error. To quantify loss sharpness, we introduce the concept of Local Robust Radius, measuring the smallest perturbation required to flip a prediction and establishing a theoretical link between local stability and generalization. Building on this perspective, we further observe a continual decrease in the robust radius during training, indicating weakened local stability and an increasingly sharp loss landscape that gives rise to MSF. To jointly solve the MSF phenomenon and the intractability of radius, we develop an energy-based formulation that is theoretically proven to be monotonically correlated with the robust radius, offering a tractable and principled objective for modeling flatness and stability. Building on these insights, we propose an energy-driven generative augmentation framework (E2A) that leverages energy-guided latent perturbations to generate pseudo-OOD samples and enhance model generalization. Extensive experiments across multiple benchmarks demonstrate that E2A consistently improves graph OOD generalization, outperforming state-of-the-art baselines. Code is available at https://github.com/anders1123/E2A
Hypergraphs are the natural description of higher-order interactions among objects, widely applied in social network analysis, cross-modal retrieval, etc. Hypergraph Neural Networks (HGNNs) have become the dominant solution for learning on hypergraphs. Traditional HGNNs are extended from message passing graph neural networks, following the homophily assumption, and thus struggle with the prevalent heterophilic hypergraphs that call for long-range dependence modeling. Existing solutions enlarge the message flow through the hypergraph bottleneck, mitigating the oversquashing issue and capturing long-range dependence. However, they often accelerate the loss of representation distinguishability in the repeated aggregations, leading to oversmoothing. This dilemma motivates an interesting question: Can we develop a unified mechanism that is agnostic to both homophilic and heterophilic hypergraphs? In this paper, we achieve the best of both worlds through the lens of Riemannian geometry, which provides the potential to adjust the message passing behavior in different regions. The key insight lies in the connection between oversquashing and hypergraph bottleneck within the framework of Riemannian manifold heat flow. Building on this, we propose the novel idea of locally adapting the bottlenecks of different subhypergraphs. The core innovation of the proposed mechanism is the design of an adaptive local (heat) exchanger. Specifically, it captures the rich long-range dependencies via the Robin condition, and preserves the representation distinguishability via source terms, thereby enabling heterophily-agnostic message passing with theoretical guarantees. Based on this theoretical foundation, we present a novel Heat-Exchanger with Adaptive Locality for Hypergraph Neural Network (HealHGNN), designed as a node-hyperedge bidirectional systems with linear complexity in the number of nodes and hyperedges. Extensive experiments on both homophilic and heterophilic cases show that HealHGNN achieves the state-of-the-art performance.
This paper investigates a novel and critical problem of Graph Cross-Domain Continual Fine-Tuning, which aims to adapt a large pre-trained Graph Foundation Model across diverse domains. Existing continual graph learning methods are mostly limited to incremental settings within only a single domain, and are typically trained from scratch. As a result, they fail to handle cross-domain shifts effectively, suffer from severe forgetting, and lack transferability. To address these challenges, we present G-CORMoL, Graph Continual Fine-tuning with Orthogonal, Router-driven Mixture of LoRA experts. G-CORMoL achieves effective adaptation while preserving prior knowledge by enforcing mathematical orthogonality between expert LoRA adapters, thereby eliminating interference across tasks. It further supports cross-domain knowledge transfer through a symmetric dual-driven routing mechanism that learns a global composition policy over all learned LoRA experts. In addition, it promotes expert specialization via a contrastive objective with theoretical guarantees. Extensive experiments on different cross-domain task orders demonstrate that G-CORMoL achieves robust state-of-the-art performance, not only preventing catastrophic forgetting but also leveraging accumulated knowledge to enable positive transfer.
A significant challenge in developing graph foundational models is to achieve universality and generalizability across general graphs, beyond text-attribute graphs. Unfortunately, most graph neural networks are designed for specific applications in particular fields and are difficult to generalize across different graphs. The most critical problem is the lack of Transfer Invariant Metadata (TIM) for graphs, akin to pixels for images and vocabularies for text, which prevents the development of graph foundational models. TIM is particularly problematic, albeit critical, on graph nodes with semantically variable features, making it difficult to transfer knowledge across graphs. Here, we analyze TIM and propose a theoretical approach to mining TIM in node features. It extracts semantically consistent information across domains and unifies data space with relatively low information loss. We then introduce a Transfer-Invariant Graph (TIG) foundational model to transform features of different dimensions into a unified structural representation. This transformation can effectively learn and extract TIM and intrinsic graph knowledge by self-supervised learning. We conducted extensive experiments, and the results showed that TIG even outperformed some models trained on data from targeted domains. In one-shot cross-domain scenarios, TIG achieved high accuracy with less training data and without any prompt tuning.
Graph Foundation Models (GFMs) have emerged as a frontier in graph learning, which are expected to deliver transferable representations across diverse tasks. However, GFMs remain constrained by in-memory bottlenecks: they attempt to encode knowledge into model parameters, which limits semantic capacity, introduces heavy lossy compression with conflicts, and entangles graph representation with the knowledge in ways that hinder efficient adaptation, undermining scalability and interpretability. In this work,we propose RAG-GFM, a Retrieval-Augmented Generation aided Graph Foundation Model that offloads knowledge from parameters and complements parameterized learning. To externalize graph knowledge, we build a dual-modal unified retrieval module, where a semantic store from prefix-structured text and a structural store from centrality-based motif. To preserve heterogeneous information, we design a dual-view alignment objective that contrasts both modalities to capture both content and relational patterns. To enable efficient downstream adaptation, we perform in-context augmentation to enrich supporting instances with retrieved texts and motifs as contextual evidence. Extensive experiments on five benchmark graph datasets demonstrate that RAG-GFM consistently outperforms 13 state-of-the-art baselines in both cross-domain node and graph classification, achieving superior effectiveness and efficiency.
Foundation models are at the forefront of artificial intelligence. A tokenizer, converting the raw input into discrete representations that the model can understand, plays an important role to the success of foundation models. Unlike the text tokenizer that is well studied in large language models, graph tokenizer is still at its early stage, facing the challenges of tackling the non-Euclidean structures and capturing the structural semantics. How to design a graph tokenizer for structural knowledge transfer? To this end, we propose a Riemannian Graph Tokenizer (RGT ) that bridges the structural knowledge and quantized representations to support cross-domain structural knowledge transfer. The connection is established by Riemannian geometry. Specifically, we first define the geometric vocabulary (trees, cycles and sequences), which captures fundamental structural patterns and reflects the intrinsic geometry of graph. Second, we construct a Riemannian quantizer with Riemannian Straight-Through Estimator to tokenise graph structures across multiple domains into discrete tokens. To ensure consistency and transferability across diverse geometric spaces, RGT further incorporates a geometry-aligned decoder that projects manifold-specific tokens into a unified tangent space. The theoretical analysis and geometric interpretations are provided to support the effectiveness of our proposed method. Extensive experiments across diverse datasets demonstrate that RGT significantly enhances structural knowledge transferability across graph domains.
Graph anomaly detection (GAD) identifies nodes, edges, and subgraphs that deviate from normal patterns, playing a key role in cybersecurity, finance, and web technologies. However, existing deep learning paradigms trained on fixed datasets struggle to generalize to unseen graphs, while current graph foundation models (GFMs) face inherent challenges in anomaly detection. Their generalization objectives conflict with the rarity and divergence of anomalies, and existing attempts to develop GFMs for GAD mainly fall into two directions: single-task cross-domain models and domain-specific multi-task models. The former transfers knowledge across domains but fails to capture cross-level dependencies, whereas the latter handles multiple tasks jointly but lacks cross-domain adaptability. Both rely heavily on labeled data, limiting scalability. These limitations highlight the need for a unified unsupervised graph foundation model capable of generalizing across domains and detecting anomalies at multiple levels. We propose GFM-UAD, a Graph Foundation Model for Unified Anomaly Detection. It is pretrained on a single dataset and performs multi-level anomaly detection across unlabeled domains. To address anomaly scarcity, GFM-UAD extracts high-confidence normal samples and synthesizes diverse anomalies through adversarial generation. By freezing the backbone and fine-tuning the detection head with distribution alignment and message passing, the model adapts flexibly to unseen environments without supervision. Experiments across multiple datasets demonstrate that GFM-UAD achieves superior accuracy and generalization, establishing a foundation for future unsupervised graph foundation models in anomaly detection.
Recently, Graph Foundation Models (GFMs) have emerged as a central focus in the field of graph learning due to their strong generalizability to various unseen graphs. However, existing GFMs typically work under the homophily assumption, and the exploration of universality on heterophilic graphs is still in its early stages. In fact, even in homophilic graphs, there exists limited yet informative heterophilic information that is not fully exploited by current GFMs. Moreover, due to the requirement for universality, the heterophily issue faced by GFMs is more challenging than in classical graph learning, as it requires training a single model to adapt to varying structures, features, and tasks. Classic heterophilic graph learning methods primarily based on the node-level homophily or heterophily. However, we highlight that homophily and heterophily exist not only at the node semantic level, but also at a finer granularity across individual feature dimensions. This finding enables GFMs to adapt to heterophilic graphs and better utilize the small amount of heterophilic information in homophilic graphs. Based on this, we propose Topology-aware Feature Sorting Graph Foundation Model (TFSGFM), which employs a feature-level topology-aware sorting strategy and a dual-channel graph neural network framework, enabling unified modeling of both feature and structure. Extensive experiments demonstrate the strong generalizability of TFSGFM. The source code is available at https://github.com/hedongxiao-tju/TFSGFM.
The remarkable success of large language models (LLMs) has motivated researchers to adapt them as universal predictors for various graph-related tasks, with the ultimate goal of developing a graph foundation model that generalizes diverse scenarios. The key challenge is to align graph data with language spaces so that LLMs can better comprehend graphs. As a popular paradigm, Graph-Tokenizing LLMs (GTokenLLMs) encode complex structures and lengthy texts into a graph token sequence, and then align them with text tokens via language instructions tuning. Despite their initial success, our information-theoretic analysis reveals that existing GTokenLLMs rely solely on text supervision from language instructions, which achieve only implicit graph–text alignment, resulting in a text-dominant bias that underutilizes graph context. To overcome this limitation, we first prove that the alignment objective is upper-bounded by the mutual information between the input graphs and their hidden representations in the LLM, which motivates us to improve this upper bound to achieve better alignment. To this end, we further propose a reconstructive graph instruction tuning pipeline, RGLM. Our key idea is to reconstruct the graph information from the LLM's graph token outputs, explicitly incorporating graph supervision to constrain the alignment process. Technically, we embody RGLM by exploring three distinct variants from two complementary perspectives: RGLM-Decoder from the input space; RGLM-Similarizer and RGLM-Denoiser from the latent space. Additionally, we theoretically analyze the alignment effectiveness of each variant. Extensive experiments on various benchmarks and task scenarios validate the effectiveness of the proposed RGLM, paving the way for new directions in GTokenLLMs' alignment research.
Digital labor markets are increasingly used to procure diverse forms of human input, from data annotation to food delivery. A central concern in such markets is the ability of platforms to suppress wages by exploiting the abundance of low-cost labor. To better understand labor outcomes, we introduce a novel posted-price procurement model with coverage objectives. In our model a platform seeks to acquire a set of M tasks while minimizing wait time, as well as total spending. Workers are sampled from a population and complete a task if the posted price is higher than their estimated cost of labor. First, we show that with a simple pricing strategy the platform can cover all categories in time O(M), while paying only a O(log(M)/M) fraction of the total cost of labor by exploiting high worker uncertainty about their costs. Then, we study the impact of collective action to prevent the exploitation of workers. We show how a small, strategically chosen coalition of workers that commits to a price floor forces the platform's total spending from logarithmic to linear in M, substantially reducing the power of the platform to suppress wages. In contrast, a randomly sampled coalition of equal size remains largely ineffective. We complement our theory with synthetic experiments that demonstrate the benefit of targeted recruitment for collective action across different market regimes. More broadly, our results provide a theoretical foundation for understanding how to organize collective bargaining for promoting welfare in digital labor markets.