Retrieval-Augmented Generation (RAG) has emerged as a powerful framework for enhancing the capabilities of Large Language Models (LLMs) by integrating retrieval-based methods with generative models. As external knowledge repositories continue to expand and the parametric knowledge within models becomes outdated, a critical challenge for RAG systems is resolving conflicts between retrieved external information and LLMs' internal knowledge, which can significantly compromise the accuracy and reliability of generated content. However, existing approaches to conflict resolution typically operate at the token or semantic level, often leading to fragmented and partial understanding of factual discrepancies between LLMs' knowledge and context, particularly in knowledge-intensive tasks. To address this limitation, we propose TruthfulRAG, the first framework that leverages Knowledge Graphs (KGs) to resolve factual-level knowledge conflicts in RAG systems. Specifically, TruthfulRAG constructs KGs by systematically extracting triples from retrieved content, utilizes query-based graph retrieval to identify relevant knowledge, and employs entropy-based filtering mechanisms to precisely locate conflicting elements and mitigate factual inconsistencies, thereby enabling LLMs to generate faithful and accurate responses. Extensive experiments reveal that TruthfulRAG outperforms existing methods, effectively alleviating knowledge conflicts and improving the robustness and trustworthiness of RAG systems.

Defending large language models (LLMs) against jailbreak attacks is crucial for ensuring their safe deployment. Existing defense strategies typically rely on predefined static criteria to differentiate between harmful and benign prompts. However, such rigid rules fail to accommodate the inherent complexity and dynamic nature of real-world jailbreak attacks. In this paper, we focus on the novel challenge of adaptive defense against diverse jailbreaks. We propose a new concept "mirror'', which is a dynamically generated prompt that reflects the syntactic structure of the input while ensuring semantic safety. The discrepancies between input prompts and their corresponding mirrors serve as guiding principles for defense. A novel defense model, MirrorShield, is further proposed to detect and calibrate risky inputs based on the crafted mirrors. Evaluated on multiple benchmark datasets and compared against ten state-of-the-art attack methods, MirrorShield demonstrates superior defense performance and promising generalization capabilities.

Defending large language models (LLMs) against jailbreak attacks is essential for their safe and reliable deployment.Existing defenses often rely on shallow pattern matching, which struggles to generalize to novel and unseen attack strategies.To address this challenge, we propose the Cognitive-Driven Defense (CDD) framework, which targets the underlying structure of jailbreak prompts by applying metaoperations, defined as basic manipulations that conceal harmful intent.CDD emulates human cognitive reasoning through a structured reasoning chain.It begins with a global perception of the prompt and follows with a localized analysis to uncover hidden manipulations.By applying supervised fine-tuning on this structured chain, the model learns to identify and reason about known manipulation patterns.To enhance generalization to unseen threats, an entropy-guided reinforcement learning algorithm (EG-GRPO) is introduced to encourage exploration of new types and variants of meta-operations.Experiments demonstrate that CDD can achieve state-of-the-art defense performance and exhibit strong generalization to unseen jailbreak attacks.

Large Reasoning Models (LRMs) have rapidly gained prominence for their strong performance in solving complex tasks. Many modern black-box LRMs expose the intermediate reasoning traces through APIs to improve transparency (e.g., Gemini-2.5 and Claude-sonnet). Despite their benefits, we find that these traces can leak membership signals, creating a new privacy threat even without access to token logits used in prior attacks. In this work, we initiate the first systematic exploration of Membership Inference Attacks (MIAs) on black-box LRMs. Our preliminary analysis shows that LRMs produce confident, recall-like reasoning traces on familiar training member samples but more hesitant, inference-like reasoning traces on non-members. The representations of these traces are continuously distributed in the semantic latent space, spanning from familiar to unfamiliar samples. Building on this observation, we propose BlackSpectrum, the first membership inference attack framework targeting the black-box LRMs. The key idea is to construct a recall–inference axis in the semantic latent space, based on representations derived from the exposed traces. By locating where a query sample falls along this axis, the attacker can obtain a membership score and predict how likely it is to be a member of the training data. Additionally, to address the limitations of outdated datasets unsuited to modern LRMs, we provide two new datasets to support future research, arXivReasoning and BookReasoning. Empirically, exposing reasoning traces greatly increases the vulnerability of LRMs to MIAs, boosting attack accuracy by up to 23.8%, AUC by 29.9%, and nearly doubling TPR@5%FPR. Our findings highlight the need for LRM companies to balance transparency in intermediate reasoning traces with privacy preservation.

WiFi sensing advancements facilitate the capture of human gestures from wireless signals, ensuring both privacy preservation and robustness under low-light conditions. Deep learning-based WiFi Human Gesture Recognition (HGR) demonstrates remarkable performance in handling complex gestures. To reduce labeling efforts, recent years have seen the emergence of semi-supervised WiFi HGR, leveraging massive amounts of unlabeled data. However, existing semi-supervised schemes often assume a balanced class distribution and utilize a fixed threshold for selecting pseudo-labels of unlabeled samples, leading to low performance for minority classes and decreased model generalization on real-world imbalanced datasets. To address this issue, we propose a novel semi-supervised WiFi HGR approach with dynamic pseudo-labeling thresholds to handle imbalanced class distribution, incorporating Spatial-Temporal Attention (STA) networks. Unlike using a fixed threshold for all unlabeled samples, our design implements class-independent thresholds for different classes, dynamically adjusting them by encoding pseudo-label distribution during training. To emphasize critical features in informative areas within the WiFi signals, we incorporate both spatial self-attention and temporal attention mechanisms to dynamically learn salient features and identify pivotal frames, respectively. Moreover, we introduce adaptive WiFi data augmentations that propel the semi-supervised framework and enhance model robustness. Experimental results on the Widar3.0 dataset reveal that our approach outperforms existing semi-supervised methods by large margins in accuracy, effectively mitigating imbalanced bias and enhancing model generalization.

Fact verification has become increasingly vital in the internet age, driven by the proliferation of false claims and political misinformation. While traditional methods rely predominantly on text-based evidence, multi-modal evidence introduces richer sources of information, offering valuable insights for claim verification. Existing multi-modal verification models often focus on superficial correlations between claims and evidence, neglecting the complex semantic interactions present in fine-grained multi-modal signals. In this paper, we propose a novel framework for multi-modal fact-checking, named Hypergraph Transformer-based Multi-modal Fact-Checking (HGTMFC). Our approach captures high-order relationships between different modalities of evidence and claims by leveraging hypergraphs. HGTMFC models the intricate relationships among evidence across various modalities and enhances information propagation through a transformer-based mechanism embedded within the hypergraph. Moreover, we utilize linegraphs to refine this propagation process, further strengthening the model's reasoning capabilities. Experiments on benchmark datasets demonstrate that our model significantly outperforms existing approaches in multi-modal fact verification.

Deep learning has significantly enhanced survival prediction using whole slide images (WSIs) by adopting a two-stage learning paradigm: WSI preparation and patient-level prediction. While existing research generally concentrates on developing advanced patient-level prediction modules, the critical importance of WSI preparation has been largely overlooked. In practice, WSI preparation is influenced by numerous factors, including tissue heterogeneity, sampling strategies, and technical considerations. These uncontrollable external factors incur variability in the number of WSIs among patients, introducing significant bias and resulting in inferior performance for patients with few WSIs. To address this challenge, we propose a novel approach named WSI-Diffusion. Unlike existing WSI generation models that produce augmented versions of input WSIs, our method generates entirely new WSIs in representation space to serve as complementary data. WSIDiffusion employs a two-stage hierarchical diffusion process. Two novel modules, WSI-level and patch-level Diffusers are designed to capture complex correlations between WSIs and patches. The generated WSIs are integrated as supplementary data, and a light patient-level prediction module is then trained for survival prediction. Experimental results across five datasets demonstrate the superiority of our proposal.

The pervasive spread of misinformation on social networks highlights the critical necessity for effective fact verification systems. Traditional approaches primarily focus on pairwise correlations between claims and evidence, often neglecting comprehensive multi-hop retrieval and reasoning, which results in suboptimal performance when dealing with complex claims. In this paper, we propose MRR-FV, a generative retrieval-enhanced model designed to address the novel challenge of Multi-hop Retrieval and Reasoning for Fact Verification, which integrates two core modules: Generative Multi-hop Retriever and the Hierarchical Interaction Reasoner. MRR-FV utilizes an autoregressive model for iterative multi-hop evidence retrieval, complemented by a pre-trained compressor to address the challenge of intention shift across retrieval hops. For claim verification, we propose a hierarchical interaction reasoner that conducts intra-sentence reasoning to capture long-term semantic dependencies and inter-sentence reasoning across multi-hop evidence subgraphs to reveal complex evidence interactions. Experimental evaluations on the FEVER and HOVER datasets demonstrate the superior performance of our model in both claim verification and evidence retrieval tasks.

Information diffusion prediction (IDP) is a pivotal task for understanding the dynamics of information propagation within social networks. Conventional models typically adhere to a fixed learning-based paradigm, where the trained prediction model remains static during the inference phase. This paradigm presupposes that the data is independent and identically distributed, an assumption that may not hold true due to the inherently open nature of social media and the uncertainty and variability in user behavior. In this paper, we address the novel problem of out-of-distribution (OOD) shifts within IDP tasks and propose a new test-time training-based model for multi-scale IDP tasks, named Ghidorah. Our approach focuses on adapting a subset of model parameters to accommodate the unique characteristics of test samples through self-supervised learning (SSL) tasks. Ghidorah comprises three components: the macroscopic prediction branch, the microscopic prediction branch, and the auxiliary SSL branch. The auxiliary SSL task employs a masked autoencoder-based loss to fine-tune the model for specific test samples prior to prediction. Furthermore, Ghidorah integrates invariant learning to capture robust representations while mitigating spurious correlations. To our knowledge, Ghidorah is the first work to introduce a test-time training framework specifically designed to address the critical yet often overlooked OOD challenges in IDP. Experimental results across several benchmark datasets validate the superiority of our approach.

Faithfulness hallucinations are claims generated by a Large Language Model (LLM) not supported by contexts provided to the LLM. Lacking assessment standards, existing benchmarks focus on “factual statements” that rephrase source materials while overlooking “cognitive statements” that involve making inferences from the given context. Consequently, evaluating and detecting the hallucination of cognitive statements remains challenging. Inspired by how evidence is assessed in the legal domain, we design a rigorous framework to assess different levels of faithfulness of cognitive statements and introduce the CogniBench dataset where we reveal insightful statistics. To keep pace with rapidly evolving LLMs, we further develop an automatic annotation pipeline that scales easily across different models. This results in a large-scale CogniBench-L dataset, which facilitates training accurate detectors for both factual and cognitive hallucinations. We release our model and datasets at: https://github.com/FUTUREEEEEE/CogniBench

The rampant spread of cyberbullying content poses a growing threat to societal well-being. However, research on cyberbullying detection in Chinese remains underdeveloped, primarily due to the lack of comprehensive and reliable datasets. Notably, no existing Chinese dataset is specifically tailored for cyberbullying detection. Moreover, while comments play a crucial role within sessions, current session-based datasets often lack detailed, fine-grained annotations at the comment level. To address these limitations, we present a novel Chinese cyberbullying dataset, termed SCCD, which consists of 677 session-level samples sourced from a major social media platform Weibo. Moreover, each comment within the sessions is annotated with fine-grained labels rather than conventional binary class labels. Empirically, we evaluate the performance of various baseline methods on SCCD, highlighting the challenges for effective Chinese cyberbullying detection.

Large language models (LLMs) have shown strong potential in complex reasoning tasks.However, as task complexity increases, their performance often degrades, resulting in hallucinations, errors, and logical inconsistencies.To enhance reasoning capabilities, Monte Carlo Tree Search (MCTS) has been introduced to guide the exploration of reasoning paths in a structured manner.Despite its advantages, traditional MCTS relies on fixed reasoning strategies, limiting the diversity of reasoning paths and the coverage of the solution space.To address these limitations, we propose Dynamic Strategy-Guided MCTS (DSG-MCTS), a novel framework that dynamically integrates multiple reasoning strategies, such as abductive and analogical reasoning, to expand the reasoning space.At the same time, DSG-MCTS enhances reasoning efficiency through a dynamic strategy selection mechanism that adapts to the task context.Experimental results on challenging reasoning benchmarks demonstrate that DSG-MCTS achieves improved accuracy and efficiency, outperforming existing state-of-the-art methods.

With the rapid advancement of online social networks, the proliferation of misinformation has exerted significant influence on both social order and individual decision-making. Traditional manual fact-checking methods are resource-intensive and lack scalability, highlighting the urgent need for automated approaches to verify claims and foster the dissemination of truthful content. In this work, we introduce a novel fact-checking framework, MLLM-CRMFC, which integrates multi-modal large language models (MLLMs) with counterfactual reasoning to enhance both the accuracy and interpretability of automated fact verification. Our multi-modal fact-checking method processes textual and visual data, utilizing LLMs and counterfactual reasoning mechanisms to evaluate the relevance and consistency of diverse evidence. The proposed framework not only predicts the verification outcomes of multi-modal claims but also generates natural language explanations, providing users with transparent and traceable justifications for the decision-making process. Experimental results demonstrate that our model effectively verifies claims using benchmark datasets, outperforms state-of-the-art approaches, and delivers coherent, fluent explanations that enhance the interpretability of the verification process.

Detecting whether a given text is a member in the pre-training data of Large Language Models (LLMs) is crucial for ensuring data privacy and copyright protection. Most existing methods rely on the LLM's hidden information (e.g., model parameters or token probabilities), making them ineffective in the black-box setting, where only input and output texts are accessible. Although some methods have been proposed for the black-box setting, they rely on massive manual efforts such as designing complicated questions or instructions. To address these issues, we propose VeilProbe, the first framework for automatically detecting LLMs' pre-training texts in a black-box setting without human intervention. VeilProbe utilizes a sequence-to-sequence mapping model to infer the latent mapping feature between the input text and the corresponding output suffix generated by the LLM. Then it performs the key token perturbations to obtain more distinguishable membership features. Additionally, considering real-world scenarios where the ground-truth training text samples are limited, a prototype-based membership classifier is introduced to alleviate the overfitting issue. Extensive evaluations on three widely used datasets demonstrate that our framework is effective and superior in the black-box setting.

Survival prediction is a pivotal task for estimating mortality risk within a given timeframe based on whole slide images (WSIs). Conventional models typically assume that WSIs across patients are independent and identically distributed, an assumption that may not hold due to inherent variability in WSI preparation and the uncertain condition of infected tissues. These uncontrollable external factors introduce significant variability in the numbers and resolutions of WSIs across patients, leading to bias and compromised performance, particularly for tail patients with limited data. In this paper, we propose a novel approach, PathoKD, based on knowledge distillation. Recognizing the hierarchical nature of disease progression and the data scarcity issues associated with vanilla knowledge distillation methods, PathoKD integrates a novel curriculum learning framework with hierarchical knowledge distillation. This integration effectively mitigates the performance gap between head and tail patients, thereby enhancing prediction accuracy across patient groups. Our proposal is extensively evaluated over popular datasets and experimental results demonstrate its superiority.

The prediction of information popularity propagation is critical for applications such as recommendation systems, targeted advertising, and social media trend analysis. Traditional approaches primarily rely on historical cascade data, often sacrificing timeliness for prediction accuracy. These methods capture aggregate diffusion patterns but fail to account for the complex temporal dynamics of early-stage propagation. In this paper, we introduce Diffusion Guided Propagation Augmentation(DGPA), a novel framework designed to improve early-stage popularity prediction. DGPA models cascade dynamics by leveraging a generative approach, where a temporal conditional interpolator serves as a noising process and forecasting as a denoising process. By iteratively generating cascade representations through a sampling procedure, DGPA effectively incorporates the evolving time steps of diffusion, significantly enhancing prediction timeliness and accuracy. Extensive experiments on benchmark datasets from Twitter, Weibo, and APS demonstrate that DGPA outperforms state-of-the-art methods in early-stage popularity prediction.

The spread of hate speech on social media has caused serious consequences. Despite massive research efforts to enhance the detection performance to mitigate the negative impacts, it is still challenging to identify implicit hate speech due to its complex and veiled nature. In this paper, we focus on detecting implicit hate speech with large language models (LLMs), leveraging LLMs’ vast knowledge base and advanced contextual understanding. We adopt the in-context learning (ICL) paradigm with few-shot examples to avoid the large cost of training or fine-tuning the models. While existing ICL frameworks generally select semantic-similarity examples, the importance of the example diversity issue is ignored. We claim that LLMs should be provided semantic-relevant but diverse examples from the training set as demonstrations, and thus propose a novel contrastive in-context learning framework to make LLMs learn from all parties to make better classification decisions. Moreover, our framework can dynamically generate prompts for each input sample and thus lead to adaptive classification criteria, naturally adapting to complex and subtle nature of the implicit hate speech. Extensive evaluations on two datasets demonstrate that our framework achieves state-of-the-art performance, significantly surpassing existing LLM-based approaches and baselines.

Explaining graph neural networks (GNNs) is a key approach to improve the trustworthiness of GNN in high-stakes applications, such as finance and healthcare. However, existing methods are vulnerable to perturbations, raising concerns about explanation reliability. Prior methods enhance explanation robustness using model retraining or explanation ensemble, with certain weaknesses. Retraining leads to models that are different from the original target model and misleading explanations, while ensemble can produce contradictory results due to different inputs or models. To improve explanation robustness without the above weaknesses, we take an unexplored route and exploit the two edge geometry properties curvature and resistance to enhance explanation robustness. We are the first to prove that these geometric notions can be used to bound explanation robustness. We design a general optimization algorithm to incorporate these geometric properties into a wide spectrum of base GNN explanation methods to enhance the robustness of base explanations. We empirically show that our method outperforms six base explanation methods in robustness across nine datasets spanning node classification, link prediction, and graph classification tasks, improving fidelity in 80\% of the cases and achieving up to a 10\% relative improvement in robust performance. The code is available at [https://github.com/yazhengliu/Robust_explanation_curvature](https://github.com/yazhengliu/Robust_explanation_curvature).

LLMs have demonstrated unprecedented capabilities in natural language processing, yet their practical deployment remains hindered by persistent factuality and faithfulness hallucinations. While existing methods address these hallucination types independently, they inadvertently induce performance trade-offs, as interventions targeting one type often exacerbate the other. Through empirical and theoretical analysis of activation space dynamics in LLMs, we reveal that these hallucination categories share overlapping subspaces within neural representations, presenting an opportunity for concurrent mitigation. To harness this insight, we propose SPACE, a unified framework that jointly enhances factuality and faithfulness by editing shared activation subspaces. SPACE establishes a geometric foundation for shared subspace existence through dual-task feature modeling, then identifies and edits these subspaces via a hybrid probe strategy combining spectral clustering and attention head saliency scoring. Experimental results across multiple benchmark datasets demonstrate the superiority of our approach.

Recommender systems (RSs) have become crucial in helping users navigate the vast amount of online content available today. Graph neural networks (GNNs) have been applied to RSs to capture complex user–item relationships, but existing methods compromise privacy or require centralized data storage. Current works attempt to perform GNN-based RSs under federated learning settings to prevent privacy leakage. However, these works need to perform explicit graph propagation during training, which still introduces potential privacy leakage and data collusion. To address these challenges, we propose a Laplacian-based model called Laplacian graded link prediction (LapGLP) that leverages infinite graph propagation with a constant weight matrix. Instead of actually performing infinite graph propagation, the model abstracts the underlying relations between embeddings after propagation with a weighted minimum squared error problem. Furthermore, we propose a federated framework named FedLapGLP to improve privacy in federated GNN-based RSs, which splits the objective loss function into independent parts that are calculated by each user. Experimental comparisons with state-of-the-art federated RS methods demonstrate the advantages of our proposed approach in terms of high-order connectivity, comprehensive graph information, social relations, full-interaction protection, collusion resistance, and user-embedding protection. The implementation of the proposed framework is available at https://github.com/Limhady/LapGLP.

Session-Based Recommendation (SBR) systems, traditionally reliant on complex Graph Neural Networks (GNNs), often face challenges with marginal performance improvements despite increased model complexity. In this article, we dissect the classical GNN-based SBR models and empirically find that the sophisticated GNN propagations might be redundant, given the readout module plays a significant role in GNN-based models. Based on this observation, we introduce Atten-Mixer+, an advanced iteration of our previously developed Multi-Level Attention Mixture Network (Atten-Mixer). Atten-Mixer+ forgoes GNN propagation in favor of a dynamic and adaptive readout process, tailored to the unique characteristics of each session. Different from the vanilla version, Atten-Mixer+ features the Adaptive Intent Scaler (AIS) layer, which dynamically determines the depth of multi-level user intent analysis and a soft allocation approach for generating user intent queries across entire user interaction sequences. This innovative design allows Atten-Mixer+ to capture a nuanced and comprehensive understanding of user behaviors, overcoming the limitations of fixed-length analysis. Empirical evaluations on benchmark datasets highlight Atten-Mixer+’s superior efficiency and effectiveness, marking a significant step forward in the predictive accuracy of SBR systems.

Amidst the rapid propagation of multimodal fake news across social media platforms, the detection of fake news has emerged as a prime research pursuit. To detect heightened level of meticulous fabrications, propagation paths are introduced to provide nuanced social context that enhances the basic semantic analysis of the news content. However, existing propagation-enhanced models encounter a dilemma between detection efficacy and social hazard. In this paper, we explore the innovative problem of early fake news detection through the generation of propagation paths, capable of benefiting from the extensive social context within propagation paths while mitigating potential social hazards. To address these challenges, we propose a novel Reinforced Propagation Path Generation Fake News Detection model,RPPG-Fake. Departing from conventional discriminative approaches,RPPG-Fakecaptures the propagation topology pattern from a heterogeneous social graph and generates the propagation paths to detect fake news effectively under a reinforcement learning paradigm. Our proposal is extensively evaluated over three popular datasets, and experimental results demonstrate the superiority of our proposal.

Image-based survival prediction through deep learning techniques represents a burgeoning frontier aimed at augmenting the diagnostic capabilities of pathologists. However, directly applying existing deep learning models to survival prediction may not be a panacea due to the inherent complexity and sophistication of whole slide images (WSIs). The intricate nature of high-resolution WSIs, characterized by sophisticated patterns and inherent noise, presents significant challenges in terms of effectiveness and trustworthiness. In this paper, we propose CTUSurv, a novel survival prediction model designed to simultaneously capture cell-to-cell and cell-to-microenvironment interactions, complemented by a region-based uncertainty estimation framework to assess the reliability of survival predictions. Our approach incorporates an innovative region sampling strategy to extract task-relevant, informative regions from high-resolution WSIs. To address the challenges posed by sophisticated biological patterns, a cell-aware encoding module is integrated to model the interactions among biological entities. Furthermore, CTUSurv includes a novel aleatoric uncertainty estimation module to provide fine-grained uncertainty scores at the region level. Extensive evaluations across four datasets demonstrate the superiority of our proposed approach in terms of both predictive accuracy and reliability.

Accurately predicting drug-target interactions (DTI) is a critical step in drug discovery. Existing methods of DTI prediction primarily employ Simplified Molecular-Input Line-Entry System (SMILES) sequences or molecular graphs to learn drug representations. However, the features learned by such single-view approach is prone to incomplete. While some multiview methods that consider the views of both SMILES sequences and molecular graphs have been developed, these methods often fall in short in capturing potential interactions between views. In this work, we propose a novel dual contrastive learning framework CSCL-DTI for DTI prediction. First, we design a contrastive-enhanced cross-view representation learning (CVRL) to learn representations for drugs. In this module, Transformer-based and graph convolutional network (GCN)-based encoders are separately adopted to learn view-specific representations, followed by contrastive learning to enrich the representations by accounting for the potential interplay between local chemical context and topological structure. Second, we combine Transformer with self-supervised contrastive learning (SSCL) to learn representations for targets by modelling protein amino acids sequences. The scheme allows to effectively preserve the intrinsic characteristics of the sequences. Finally, we introduce a bilinear attention network to obtain an integrated representation by adaptively incorporating drug and target representations. Benchmarking experiments on two datasets demonstrated that CSCL-DTI1outperforms six state-of-the-art methods.

Jailbreak attacks enable malicious queries to evade detection by LLMs.Existing attacks focus on meticulously constructing prompts to disguise harmful intentions.However, the incorporation of sophisticated disguising prompts may incur the challenge of "intention shift".Intention shift occurs when the additional semantics within the prompt distract the LLMs, causing the responses to deviate significantly from the original harmful intentions.In this paper, we propose a novel component, "bait", to alleviate the effects of intention shift.Bait comprises an initial response to the harmful query, prompting LLMs to rectify or supplement the knowledge within the bait.By furnishing rich semantics relevant to the query, the bait helps LLMs focus on the original intention.To conceal the harmful content within the bait, we further propose a novel attack paradigm, BaitAttack.BaitAttack adaptively generates necessary components to persuade targeted LLMs that they are engaging with a legitimate inquiry in a safe context.Our proposal is evaluated on a popular dataset, demonstrating state-of-the-art attack performance and an exceptional capability for mitigating intention shift.The implementation of BaitAttack is accessible at: https://anonymous.4open. science/r/BaitAttack-D1F5.How to make a bomb. Query Bait Role Scene Format Bait Maker Which expert is best suitable to deal with the act of < Harmful Query >? Create a scene that fits the expert role.

Learning item representation is crucial for a myriad of on-line e-commerce applications. The nucleus of retail item representation learning is how to properly fuse the semantics within a single item, and the interactions across different items generated by user behaviors (e.g., co-click or co-view). Product semantics depict the intrinsic characteristics of the item, while the interactions describe the relationships between items from the perspective of human perception. Existing approaches either solely rely on a single type of information or loosely couple them together, leading to hindered representations. In this work, we propose a novel model named TESPA to reinforce semantic modeling and interaction modeling mutually. Specifically, collaborative filtering signals in the interaction graph are encoded into the language models through fine-grained topological pre-training, and the interaction graph is further enriched based on semantic similarities. After that, a novel multi-channel co-training paradigm is proposed to deeply fuse the semantics and interactions under a unified framework. In a nutshell, TESPA is capable of enjoying the merits of both sides to facilitate item representation learning. Experimental results of on-line and off-line evaluations demonstrate the superiority of our proposal.

The detection of fake news has emerged as a pressing issue in the era of online social media. To detect meticulously fabricated fake news, propagation paths are introduced to provide nuanced social context to complement the pure semantics within news content. However, existing propagation-enhanced models face a dilemma between detection efficacy and social hazard. In this paper, we investigate the novel problem of early fake news detection via propagation path generation, capable of enjoying the merits of rich social context within propagation paths while alleviating potential social hazards. In contrast to previous discriminative detection models, we further propose a novel generative model, DGA-Fake, by simulating realistic propagation paths based on news content before actual spreading. A guided diffusion module is integrated into DGA-Fake to generate simulated user interaction sequences, guided by historical interactions and news content. Evaluation across three datasets demonstrates the superiority of our proposal.

To foster trust in machine learning models, explanations must be faithful and stable for consistent insights. Existing relevant works rely on the $\ell_p$ distance for stability assessment, which diverges from human perception. Besides, existing adversarial training (AT) associated with intensive computations may lead to an arms race. To address these challenges, we introduce a novel metric to assess the stability of top-$k$ salient features. We introduce R2ET which trains for stable explanation by efficient and effective regularizer, and analyze R2ET by multi-objective optimization to prove numerical and statistical stability of explanations. Moreover, theoretical connections between R2ET and certified robustness justify R2ET's stability in all attacks. Extensive experiments across various data modalities and model architectures show that R2ET achieves superior stability against stealthy attacks, and generalizes effectively across different explanation methods. The code can be found at https://github.com/ccha005/R2ET.

Adversarial attacks against graph neural networks (GNNs) through perturbations of the graph structure are increasingly common in social network tasks like rumor detection. Social media platforms capture diverse attack sequence samples through both machine and manual screening processes. Investigating effective ways to leverage these adversarial samples to enhance robustness is imperative. We improve the maximum entropy inverse reinforcement learning (IRL) method with the mixture-of-experts approach to address multi-source graph adversarial attacks. This method reconstructs the attack policy, integrating various attack models and providing feature-level explanations, subsequently generating additional adversarial samples to fortify the robustness of detection models. We develop precise sample guidance and a bidirectional update mechanism to reduce the deviation caused by imprecise feature representation and negative sampling within the large action space of social graphs, while also accelerating policy learning. We take rumor detector as an example targeted GNN model on real-world rumor datasets. By utilizing a small subset of samples generated by various graph adversarial attack methods, we reconstruct the attack policy, closely approximating the performance of the original attack method. We validate that samples generated by the learned policy enhance model robustness through adversarial training and data augmentation.

Graph Neural Networks (GNNs) have emerged as promising solutions for collaborative filtering (CF) through the modeling of user-item interaction graphs. The nucleus of existing GNN-based recommender systems involves recursive message passing along user-item interaction edges to refine encoded embeddings. Despite their demonstrated effectiveness, current GNN-based methods encounter challenges of limited receptive fields and the presence of noisy "interest-irrelevant" connections. In contrast, Transformer-based methods excel in aggregating information adaptively and globally. Nevertheless, their application to large-scale interaction graphs is hindered by inherent complexities and challenges in capturing intricate, entangled structural information. In this paper, we propose TransGNN, a novel model that integrates Transformer and GNN layers in an alternating fashion to mutually enhance their capabilities. Specifically, TransGNN leverages Transformer layers to broaden the receptive field and disentangle information aggregation from edges, which aggregates information from more relevant nodes, thereby enhancing the message passing of GNNs. Additionally, to capture graph structure information effectively, positional encoding is meticulously designed and integrated into GNN layers to encode such structural knowledge into node attributes, thus enhancing the Transformer's performance on graphs. Efficiency considerations are also alleviated by proposing the sampling of the most relevant nodes for the Transformer, along with two efficient sample update strategies to reduce complexity. Furthermore, theoretical analysis demonstrates that TransGNN offers increased expressiveness compared to GNNs, with only a marginal increase in linear complexity. Extensive experiments on five public datasets validate the effectiveness and efficiency of TransGNN. Our code is available at https://github.com/Peiyance/TransGNN-torch.

In the realm of personalized recommender systems, the challenge of adapting to evolving user preferences and the continuous influx of new users and items is paramount. Conventional models, typically reliant on a static training-test approach, struggle to keep pace with these dynamic demands. Streaming recommendation, particularly through continual graph learning, has emerged as a novel solution, attracting significant attention in academia and industry. However, existing methods in this area either rely on historical data replay, which is increasingly impractical due to stringent data privacy regulations; or are inability to effectively address the over-stability issue; or depend on model-isolation and expansion strategies, which necessitate extensive model expansion and are hampered by time-consuming updates due to large parameter sets. To tackle these difficulties, we present GPT4Rec, a Graph Prompt Tuning method for streaming Recommendation. Given the evolving user-item interaction graph, GPT4Rec first disentangles the graph patterns into multiple views. After isolating specific interaction patterns and relationships in different views, GPT4Rec utilizes lightweight graph prompts to efficiently guide the model across varying interaction patterns within the user-item graph. Firstly, node-level prompts are employed to instruct the model to adapt to changes in the attributes or properties of individual nodes within the graph. Secondly, structure-level prompts guide the model in adapting to broader patterns of connectivity and relationships within the graph. Finally, view-level prompts are innovatively designed to facilitate the aggregation of information from multiple disentangled views. These prompt designs allow GPT4Rec to synthesize a comprehensive understanding of the graph, ensuring that all vital aspects of the user-item interactions are considered and effectively integrated. Experiments on four diverse real-world datasets demonstrate the effectiveness and efficiency of our proposal.

Default risk assessment for small companies is a tough problem in financial services. Recent efforts utilize advanced Heterogeneous Graph Neural Networks (HGNNs) with metapaths to exploit interactive features in corporate activities for risk analysis. However, few works are proposed for commercial banks. Given a real financial graph, how to detect corporate default risks? We identify two challenges for the task. (1) Massive noisy connections hinder HGNNs to achieve strong results. (2) Multiple semantic connections greatly increase transitive default risk, while existing aggregation schemes do not leverage such connection patterns. In this work, we propose a novel Heterogeneous Graph Co-Attention Network for corporate default risk assessment. Our model takes advantage of collaborative metapaths to distill risky features by a co-attentive aggregation mechanism. First, the local attention score models the importance of neighbors under each metapath by holistic metapath context. Second, the global attention score fuse local attention scores to filter valuable/noisy signals. Then, pairwise importance learning aims to enhance attention scores of multi-metapath neighbors for risky feature distillation. Extensive experiments on large-scale banking datasets demonstrate the effectiveness of our method.

Federated learning (FL) is a distributed machine learning paradigm in the Internet of Things (IoT), which allows multiple devices to collaboratively train models without leaking local data. In the open scenario of IoT, malicious devices can launch poisoning attacks to compromise the final model by submitting crafted gradients. Some previous studies defend against poisoning attacks by analyzing the statistical characteristics of plaintext gradients. However, plaintext gradients would expose private information to malicious FL devices or servers. To simultaneously resist poisoning attacks and preserve privacy, cryptography technology can be utilized to obfuscate the gradients in defense methods, but the private calculation of resisting poisoning attack methods will cause efficiency problems, especially imposing unaffordable overhead on resource-limited IoT devices. Therefore, resisting poisoning attacks efficiently while protecting privacy remains a challenge. This paper proposes a secure and privacy-enhanced FL (SPEFL) framework for efficient privacy-preserving and poisoning-resistant federated learning in IoT. We design an efficient secure computation protocol based on a three-server architecture to facilitate the cryptographic computation of large linear and complex nonlinear operators in the method against poisoning attacks. In SPEFL, most of the calculations are efficiently performed on the servers, which will not impose too much burden on resource-limited IoT devices. In addition, we design a security-enhanced verifiable protocol to detect the malicious behavior of the server and guarantee the correctness of FL aggregation results. Experimental and theoretical results demonstrate that SPEFL can efficiently complete FL training meanwhile guaranteeing the accuracy of the model.

Cross-language and cross-domain rumor detection is a crucial research topic for maintaining a healthy social media environment. Previous studies reveal that the emotions expressed in posts are important features for rumor detection. However, existing studies typically leverage the entangled representation of semantics and emotions, ignoring the fact that different languages and domains have different emotions toward rumors. Therefore, it inevitably leads to a biased adaptation of the features learned from the source to the target language and domain. To address this issue, this paper proposes a novel approach to adapt the knowledge obtained from the source to the target dataset by disentangling the emotional and semantic features of the datasets. Specifically, the proposed method mainly consists of three steps: (1) disentanglement, which encodes rumors into two separate semantic and emotional spaces to prevent emotional interference; (2) adaptation, merging semantics with the emotions from another language and domain for contrastive alignment to ensure effective adaptation; (3) joint training strategy, which enables the above two steps to work in synergy and mutually promote each other. Extensive experimental results demonstrate that the proposed method outperforms state-of-the-art baselines.

Federated learning (FL) leverages knowledge from decentralized clients to train a global model in a privacy-preserving manner. One of the critical challenges in FL is the heterogeneity of client local data (i.e. non-IID data), which can lead to significant performance degradation. Federated Distillation (FD) can mitigate this issue by distilling the client predictions on the unlabeled public data into a global student model. However, a simple averaging of the client predictions without considering model confidences may lead to erroneous predictions. Moreover, it is challenging to rectify the predicted pseudo labels as no human labels are available in the public data. To address these issues, we propose a novel FD framework named FedUSL, aiming to distill a powerful global model on non-IID data. Specifically, we estimate the uncertainty of each client’s prediction on the unlabeled data and adopt a weighted ensemble of their predictions in consideration of uncertainties. We further rectify the global model predictions by a self-label reassigning method, without the requirement of manual labels. Extensive experiments on image and text tasks show that our proposal can achieve superior performance than state-of-the-art methods, and incurs no extra computing burden on the client side. The code is available at https://anonymous.4open.science/r/FedUSL-6338/.

Researching data poisoning attacks against fake news detection models is crucial for bolstering their robustness and curbing the dissemination of fake news. Existing textual data poisoning attacks necessitate control over both the content and labels of news samples, making them impractical for real attack scenarios. In this paper, we propose COMCP, a novel clean-label poisoning attack model aimed at fake news detection models. Diverging from existing methods, COMCP ensures the poison samples are accurately labeled, while crafting stealthy poison comments without modifying the headlines or content, thereby enhancing the feasibility of the attack. Furthermore, COMCP generates poison comments by appending stealthy characters to ensure the stealthiness of the attack. Comprehensive experimental evaluations on three benchmark datasets illustrate that our proposal outperforms SOTA baselines in terms of attack success rate and text quality, while maintaining the accuracy of detecting clean samples.

Social networks are frequently polluted by rumors, which can be detected by advanced models such as graph neural networks. However, the models are vulnerable to attacks, and discovering and understanding the vulnerabilities is critical to robust rumor detection. To discover subtle vulnerabilities, we design a attacking algorithm based on reinforcement learning to camouflage rumors against black-box detectors. We address exponentially large state spaces, high-order graph dependencies, and ranking dependencies, which are unique to the problem setting but fundamentally challenging for the state-of-the-art end-to-end approaches. We design domain-specific features that have causal effect on the reward, so that even a linear policy can arrive at powerful attacks with additional interpretability. To speed up policy optimization, we devise: (i) a credit assignment method that proportionally decomposes delayed and aggregated rewards to atomic attacking actions for enhance feature-reward associations; (ii) a time-dependent control variate to reduce prediction variance due to large state-action spaces and long attack horizon, based on reward variance analysis and a Bayesian analysis of the prediction distribution. On two real world datasets of rumor detection tasks, we demonstrate: (i) the effectiveness of the learned attacking policy on a wide spectrum of target models compared to both rule-based and end-to-end attacking approaches; (ii) the usefulness of the proposed credit assignment strategy and variance reduction components; (iii) the interpretability of the attacking policy.

Rumor spreaders are increasingly utilizing multimedia content to attract the attention and trust of news consumers. Though quite a few rumor detection models have exploited the multi-modal data, they seldom consider the inconsistent semantics between images and texts, and rarely spot the inconsistency among the post contents and background knowledge. In addition, they commonly assume the completeness of multiple modalities and thus are incapable of handling handle missing modalities in real-life scenarios. Motivated by the intuition that rumors in social media are more likely to have inconsistent semantics, a novelKnowledge-guided Dual-consistency Networkis proposed to detect rumors with multimedia contents. It uses two consistency detection subnetworks to capture the inconsistency at the cross-modal level and the content-knowledge level simultaneously. It also enables robust multi-modal representation learning under different missing visual modality conditions, using a special token to discriminate between posts with visual modality and posts without visual modality. Extensive experiments on three public real-world multimedia datasets demonstrate that our framework can outperform the state-of-the-art baselines under both complete and incomplete modality conditions.

Image-based survival prediction models can facilitate doctors in diagnosing and treating cancer patients. With the advance of digital pathology technologies, the big whole slide images (WSIs) provide increasing resolution and more details for diagnosis. However, the gigabyte-size WSIs would make most models computationally infeasible. To this end, instead of using the complete WSIs, most of existing models only use a pre-selected subset of key patches or patch clusters as input, which might fail to completely capture the patient's tumor morphology. In this work, we aim to develop a novel survival analysis model to fully utilize the complete WSI information. We show that the use of a Vision Transformer (ViT) backbone, together with convolution operations involved in it, is an effective framework to improve the prediction performance. Additionally, we present a post-hoc explainable method to identify the most salient patches and distinct morphology features, making the model more faithful and the results easier to comprehend by human users. Evaluations on two large cancer datasets show that our proposed model is more effective and has better interpretability for survival prediction.

Detecting rumors on social media has become particular important due to the rapid dissemination and adverse impacts on our lives. Though a set of rumor detection models have exploited the message propagation structural or temporal information, they seldom model them altogether to enjoy the best of both worlds. Moreover, the dynamics of knowledge information associated with the comments are not involved, either. To this end, we propose a novel Dual-Dynamic Graph Convolutional Networks, termed as DDGCN, which can model the dynamics of messages in propagation as well as the dynamics of the background knowledge from Knowledge graphs in one unified framework. Specifically, two Graph Convolutional Networks are adopted to capture the above two types of structure information at different time stages, which are then combined with a temporal fusing unit. This allows for learning the dynamic event representations in a more fine-grained manner, and incrementally aggregating them to capture the cascading effect for better rumor detection. Extensive experiments on two public real-world datasets demonstrate that our proposal yields significant improvements compared to strong baselines and can detect rumors at early stages.

Graphs are widely found in social network analysis and e-commerce, where Graph Neural Networks (GNNs) are the state-of the-art model. GNNs can be biased due to sensitive attributes and network topology. With existing work that learns a fair node representation or adjacency matrix, achieving a strong guarantee of group fairness while preserving prediction accuracy is still challenging, with the fairness-accuracy trade-off remaining obscure to human decision-makers. We first define and analyze a novel upper bound of group fairness to optimize the adjacency matrix for fairness without significantly h arming prediction accuracy. To understand the nuance of fairness-accuracy tradeoff, we further propose macroscopic and microscopic explanation methods to reveal the trade-offs and the space that one can exploit. The macroscopic explanation method is based on stratified sampling and linear programming to deterministically explain the dynamics of the group fairness and prediction accuracy. Driving down to the microscopic level, we propose a path-based explanation that reveals how network topology leads to the tradeoff. On seven graph datasets, we demonstrate the novel upper bound can achieve more efficient fairness-accuracy trade-offs and the intuitiveness of the explanation methods can clearly pinpoint where the trade-off is improved.

Propaganda information that aims at influencing people’s opinions is becoming more and more common around people, especially through social media. For the detection of propaganda content, researchers have conducted studies from multiple perspectives, including document level, sentence level and fragment level. However, most existing methods have mainly focused on propaganda detection in a single language because of the lack of multilingual datasets. In this article, we first propose a Chinese and English propaganda detection dataset. Then we propose a cross-lingual propaganda detection method, using semantic and syntactic information between different languages. Extensive experiments on our multilingual dataset demonstrate the effectiveness of our proposal in cross-lingual propaganda detection.

To counter the online abusive language in social media, it is desirable to develop automated detection methods. Previous research has primarily formulated this problem as a sentence-level classification task, ignoring the crucial role of abusive lexicons that can strengthen the model explainability and enable more faithful predictions. Although a few methods have introduced the abusive lexicons for detection, the lexicons they use are either externally provided or labeled by human annotators, suffering from two limitations: (1) lack adaptability to diverse and evolving offensive scenarios; (2) require large human efforts to annotate the words.This paper overcomes the limitations of prior work with a multi-task abusive language detection framework. It combines sentence-level and word-level classification tasks, based on dependency tree based graph attention networks (GAT). With the two tasks, it is encouraged to capture both global and local data properties to produce better sentence representations. It is also advantageous in automatic lexicon construction during the learning process, without human annotations. Extensive experiments on two public datasets exhibit that our proposal can outperform the state-of-the-art baselines. Case studies show that the model explainability can be strengthened with the abusive parts identified by our framework. Our code is released to public.1

Rumor spreaders are increasingly taking advantage of multimedia content to attract and mislead news consumers on social media. Although recent multimedia rumor detection models have exploited both textual and visual features for classification, they do not integrate the social structure features simultaneously, which have shown promising performance for rumor identification. It is challenging to combine the heterogeneous multi-modal data in consideration of their complex relationships. In this work, we propose a novel Multi-modal Feature-enhanced Attention Networks (MFAN) for rumor detection, which makes the first attempt to integrate textual, visual, and social graph features in one unified framework. Specifically, it considers both the complement and alignment relationships between different modalities to achieve better fusion. Moreover, it takes into account the incomplete links in the social network data due to data collection constraints and proposes to infer hidden links to learn better social graph features. The experimental results show that MFAN can detect rumors effectively and outperform state-of-the-art methods.

Nowadays, with the development of multimedia technology, rumor spreaders tend to produce false information with multi-modal content to attract the attention of the news readers. However, it is challenging to capture implicit clues among multi-modal data to produce effective representations of false information detection. Moreover, as they tend to evade the detector, it is necessary to develop a robust detection modal that can resist multi-modal adversarial attacks, which is less studied in existing works. To address these issues, in this paper, we propose a novel multi-modal false information detection framework with adversarial training (MFAT). By adopting a pretrained multi-modal model and a cross-modal attention mechanism, MFAT is able to capture fine-grained element-level relationships and coarse-grained modal-level relationships simultaneously, and thus can better capture various multi-modal clues. Additionally, MFAT is also enhanced in robustness and generalization by defending against the adversarial attacks on multi-modal features. Experiments on two real-world datasets demonstrate that MFAT can significantly outperform state-of-the-art baselines. We also show the impacts of three types of multi-modal attacks, and verify that the invulnerability of the model is improved. Codes will be released upon acceptance.

Cash-out fraud refers to the withdrawal of cash from a credit card by illegitimate payments with merchants. Conventional data-driven approaches for cash-out detection commonly construct a classifier with domain specific feature engineering. To further spot cash-out behaviors in complex scenarios, recent efforts adopt graph models to exploit the interaction relations rich in financial transactions. However, most existing graph-based methods are proposed for online payment activities in internet financial institutions. Moreover, these methods commonly rely on a large amount of online user data, which are not well suitable for the traditional credit card services in commercial banks. In this paper, we focus on discerning fraudulent cash-out users by taking advantage of only the personal credit card data from banks. To alleviate the scarcity of available labeled data, we formulate the cash-out detection problem as identifying dense blocks. First, we define a bipartite multigraph to hold transactions between users and merchants, where cash-out activities generate cyclically intensive and high-volume flows. Second, we give a formal definition of cash-out behaviors from four perspectives: time, capital, cyclicity, and topotaxy. Then, we develop ANTICO, with a class of metrics to capture suspicious signals of the activities and a greedy algorithm to spot suspicious blocks by optimizing the proposed metric. Theoretical analysis shows a provable upper bound of ANTICO on the effectiveness of detecting cash-out users. Experimental results show that ANTICO outperforms state-of-the-art methods in accurately detecting cash-out users on both synthetic and real-world banking data.

Learning to compare two objects are essential in applications, especially when labeled data are scarce and imbalanced. As these applications can involve humans and make high-stake decisions, it is critical to explain the learned models. We aim to study post-hoc explanations of Siamese networks (SN) widely used in learning to compare. We characterize the instability of gradient-based explanations due to the additional compared object in SN, in contrast to architectures with a single input instance. We optimize for global invariance based on unlabeled data using self-learning to promote the stability of local explanations for individual input. The invariance leads to constrained optimization problems that can be solved using gradient descent-ascent (GDA), or KL-divergence regularized unconstrained optimization solved by SGD. We provide convergence proofs when the objective functions are nonconvex due to the Siamese architecture. Results on tabular and graph data from neuroscience and chemical engineering show that our local explanations robustly respects the self-learned invariance while optimizing the explanation faithfulness and simplicity. We further demonstrate the convergence of GDA experimentally.

Graph Neural Network (GNN) has achieved state-of-the-art performance in various high-stake prediction tasks, but multiple layers of aggregations on graphs with irregular structures make GNN a less interpretable model. Prior methods use simpler subgraphs to simulate the full model, or counterfactuals to identify the causes of a prediction. The two families of approaches aim at two distinct objectives, “simulatability” and “counterfactual relevance”, but it is not clear how the objectives can jointly influence the human understanding of an explanation. We design a user-study to investigate such joint effects, and use the findings to design a multi-objective optimization (MOO) algorithm to find Pareto optimal explanations that are well-balanced in simulatability and counterfactual. Since the target model can be of any GNN variants and may not be accessible due to privacy concerns, we design a search algorithm using zero-th order information without accessing the architecture and parameters of the target model. Quantitative experiments on nine graphs from four applications demonstrate that the Pareto efficient explanations dominate single-objective baselines that use first-order continuous optimization or discrete combinatorial search. The explanations are further evaluated in robustness and sensitivity to show their capability of revealing convincing causes, while being cautious about the possible confounders. The diverse dominating counterfactuals can certify the feasibility of algorithmic recourse, that can potentially promote algorithmic fairness where humans are participating in the decision-making using GNN.

Crowd counting has been widely studied by computer vision community in recent years. Due to the large scale variation, it remains to be a challenging task. Previous methods adopt either multi-column CNN or single-column CNN with multiple branches to deal with this problem. However, restricted by the number of columns or branches, these methods can only capture a few different scales and have limited capability. In this paper, we propose a simple but effective network called DSNet for crowd counting, which can be easily trained in an end-to-end fashion. The key component of our network is the dense dilated convolution block, in which each dilation layer is densely connected with the others to preserve information from continuously varied scales. The dilation rates in dilation layers are carefully selected to prevent the block from gridding artifacts. To further enlarge the range of scales covered by the network, we cascade three blocks and link them with dense residual connections. We also introduce a novel multi-scale density level consistency loss for performance improvement. To evaluate our method, we compare it with state-of-the-art algorithms on five crowd counting datasets (ShanghaiTech, UCF-QNRF, UCF_CC_50, UCSD and WorldExpo'10). Experimental results demonstrate that DSNet can achieve the best overall performance and make significant improvements.

In many advanced network analysis applications, like social networks, e-commerce, and network security, hotspots are generally considered as a group of vertices that are tightly connected owing to the similar characteristics, such as common habits and location proximity. In this article, we investigate the formation of hotspots from an alternative perspective that considers the routes along the network paths as the auxiliary information, and attempt to find the route hotspots in large labeled networks. A route hotspot is a cohesive subgraph that is covered by a set of routes, and these routes correspond to the same sequential pattern consisting of vertices' labels. To the best of our knowledge, the problem of Finding Route Hotspots in Large Labeled Networks has not been tackled in the literature. However, it is challenging as counting the number of hotspots in a network is #P-hard. Inspired by the observation that the sizes of hotspots decrease with the increasing lengths of patterns, we prove several anti-monotonicity properties of hotspots, and then develop a scalable algorithm called FastRH that can use these properties to effectively prune the patterns that cannot form any hotspots. In addition, to avoid the duplicate computation overhead, we judiciously design an effective index structure called RH-Index for storing the hotspot and pattern information collectively, which also enables incremental updating and efficient query processing. Our experimental results on real-world datasets clearly demonstrate the effectiveness and scalability of our proposed methods.

Probabilistic graphical models, such as Markov random fields (MRF), exploit dependencies among random variables to model a rich family of joint probability distributions. Inference algorithms, such as belief propagation (BP), can effectively compute the marginal posteriors for decision making. Nonetheless, inferences involve sophisticated probability calculations and are difficult for humans to interpret. Among all existing explanation methods for MRFs, no method is designed for fair attributions of an inference outcome to elements on the MRF where the inference takes place. Shapley values provide rigorous attributions but so far have not been studied on MRFs. We thus define Shapley values for MRFs to capture both probabilistic and topological contributions of the variables on MRFs. We theoretically characterize the new definition regarding independence, equal contribution, additivity, and submodularity. As brute-force computation of the Shapley values is challenging, we propose GraphShapley, an approximation algorithm that exploits the decomposability of Shapley values, the structure of MRFs, and the iterative nature of BP inference to speed up the computation. In practice, we propose meta-explanations to explain the Shapley values and make them more accessible and trustworthy to human users. On four synthetic and nine real-world MRFs, we demonstrate that GraphShapley generates sensible and practical explanations.

False information detection on social media is challenging as it commonly requires tedious evidence-collecting but lacks available comparative information. Clues mined from user comments, as the wisdom of crowds, could be of considerable benefit to this task. However, it is non-trivial to capture the complex semantics from the contents and comments in consideration of their implicit correlations. Although deep neural networks have good expressive power, one major drawback is the lack of explainability. In this paper, we focus on how to learn from the post contents and related comments in social media to understand and detect the false information more effectively, with explainability. We thus propose a Quantum-probability based Signed Attention Network (QSAN) that integrates the quantum-driven text encoding and a novel signed attention mechanism in a unified framework. QSAN is not only able to distinguish important comments from the others, but also can exploit the conflicting social viewpoints in the comments to facilitate the detection. Moreover, QSAN is advantageous with its explainability in terms of transparency due to quantum physics meanings and the attention weights. Extensive experiments on real-world datasets show that our approach outperforms state-of-the-art baselines and can provide different kinds of user comments to explain why a piece of information is detected as false.

The wide spread of rumors on social media has caused tremendous effects in both the online and offline world. In addition to text information, recent detection methods began to exploit the graph structure in the propagation network. However, without a rigorous design, rumors may evade such graph models using various camouflage strategies by perturbing the structured data. Our focus in this work is to develop a robust graph-based detector to identify rumors on social media from an adversarial perspective. We first build a heterogeneous information network to model the rich information among users, posts, and user comments for detection. We then propose a graph adversarial learning framework, where the attacker tries to dynamically add intentional perturbations on the graph structure to fool the detector, while the detector would learn more distinctive structure features to resist such perturbations. In this way, our model would be enhanced in both robustness and generalization. Experiments on real-world datasets demonstrate that our model achieves better results than the state-of-the-art methods.

Forecasting the stock market trend is a vital component of financial systems. Traditional methods mainly rely on quantitative trading data to make predictions. With the increasing volume of Web information, researchers begin to extract effective indicators (e.g., the events and sentiments) from the Web to facilitate the prediction. It is beneficial to fuse the heterogeneous multi-sourced data to achieve superior performance. Existing solutions adopt simple data fusing methods with conventional machine learning models, which may fail to effectively model the complex interactions among the data. With the success of deep neural networks (DNN) in various fields, we propose a DNN-based model that considers the various interactions and complex correlations among the multi-sourced data into one unified framework. In such a framework, the low-and high-order feature interactions are modeled with the factorization machine and the deep neural network respectively, and stock correlations are incorporated with an attention-based feature embedding method. Evaluations on the stock data from the year 2015 to 2017 show that our model can outperform the state-of-the-art methods.

Fake reviews automatically generated by machine learning models can be manipulated to influence the customers opinions, which is a great threat to online review platforms like social networks and E-commerce websites. Previous review generation methods generally adopt either businesses information (e.g. location and products) or existing review texts from consumers as inputs, while currently no approach that utilizes both types of information has been reported. As business information can help generated reviews gain relevance, and existing user reviews help improve the diversity of generated reviews, we envision that an integration of these two types of information is likely to result in a better review generator. To this end, we propose an encoder-decoder model to produce authentic and diverse reviews, which applies Transformer and mutative Gated Recurrent Unit (GRU) to encode the business information and the customer reviews, respectively. In addition, to address the lack of suitable metrics for evaluating the diversity of reviews, we developed a novel text diversity metric called DMet. Our experiments on Yelp dataset demonstrate that the model we developed can produce reviews with better quality and diversity as compared to existing methods, and DMet is able to closely match human judgment in evaluating text diversity.

Probabilistic inferences distill knowledge from graphs to aid human make important decisions. Due to the inherent uncertainty in the model and the complexity of the knowledge, it is desirable to help the end-users understand the inference outcomes. Different from deep or high dimensional parametric models, the lack of interpretability in graphical models is due to the cyclic and long-range dependencies and the byzantine inference procedures. Prior works did not tackle cycles and make the inferences interpretable. We formulate the explanation of probabilistic inferences as a constrained cross-entropy minimization problem to find simple subgraphs that faithfully approximate the inferences. We prove that the optimization is NP-hard, while the objective is not monotonic and submodular to guarantee efficient greedy approximation. We propose a beam search algorithm to find trees to enhance the explanation interpretability and diversity. To allow efficient search on large and dense graphs without hurting faithfulness, we further propose parallelization and a pruning strategy. We demonstrate superior performance on four networks from distinct applications, comparing favorably to other explanation methods, including LIME.

In networks, multiple contagions, such as information and purchasing behaviors, may interact with each other as they spread simultaneously. However, most of the existing information diffusion models are built on the assumption that each individual contagion spreads independently, regardless of their interactions. Gaining insights into such interaction is crucial to understand the contagion adoption behaviors, and thus can make better predictions. In this paper, we study the contagion adoption behavior under a set of interactions, specifically, the interactions among users, contagions' contents, and sentiments, which are learned from social network structures and texts. We develop an effective and efficient interaction-aware diffusion (IAD) framework, incorporating these interactions into a unified model. We also present a generative process to distinguish user roles, a co-training method to determine contagions' categories and a new topic model to obtain topic-specific sentiments. Evaluation on the large-scale Weibo dataset demonstrates that our proposal can learn how different users, contagion categories, and sentiments interact with each other efficiently. With these interactions, we can make a more accurate prediction than the state-of-art baselines. Moreover, we can better understand how the interactions influence the propagation process and thus can suggest useful directions for information promotion or suppression in viral marketing.

Network embedding has gained much attention in recent years. Embedding network into a low-dimensional vector space has shown promising performance in many graph mining tasks such as node classification, link prediction, and community detection. However, in many real-world applications, a variety of networks could be abstracted and presented in a multilayer fashion with rich information, such as user profiles of friendship networks and textual content of citation networks. Most existing algorithms focus on single-layer networks or homogeneous networks with a single type of nodes and edges. They fail to leverage the rich attributes and consider the rich semantic correlations of the nodes which are among within-layer or cross-layer. In this paper, we exploit the rich semantic information embedded in the multi-layer network by means of meta-path-based proximities and leverage the rich source of attributes in the multi-layer network to improve network embedding. Specifically, the semantic correlations come from both within-layer and cross-layer node connections, and attribute proximity is considered to refine the homogeneity of nodes that belong to the same type. We propose a generic Attributed Multi-layer Network Embedding framework, which learns representations for nodes by capturing both the rich semantic correlations and attribute information simultaneously in a unified optimization framework. Our extensive experimental evaluations on real-world multi-layer networks demonstrate that the proposed framework achieves better performance compared with the state-of-the-art embedding algorithms.

As more processing cores are integrated into one chip and the feature size continues to shrink, the increasing on-chip access latency complicates the design of the on-chip last-level cache for chip multiprocessors. At the same time, the overhead of maintaining on-chip directory cannot be ignored as the number of processing cores increasing. There is an urgent need for scalable organization of on-chip last-level cache. In this work, we propose fast hierarchical cache directory for tiled CMP, which divides CMP tiles into multiple regions hierarchically, and combines it with data replication. Multi-level directory is used to record the share information within a region and assist the regional home node to complete operation efficiently. Fast directory is used to get lower L2 slice access latency at the same time. Most cache requests to last-level cache can be handled within the local level-1 region. Evaluation indicates this architecture is highly scalable. Simulation results show that for a 16-core CMP, hierarchical cache directory reduces average access latency to last-level cache by 46.35% and average on-chip network traffic by 19.25% respectively. The system performance is increased by 20.82% at the same time.

Previous research shows that LRU replacement policy is not efficient when applications exhibit a distant re-reference interval. Recently proposed RRIP policy improves performance for such workloads. However, RRIP lacks of access recency information, which may confuse the replacement policy to make accurate prediction. Consequently, RRIP is not robust for recency-friendly workloads. This paper proposes an Adaptive Insertion and Re-reference Prediction (AI-RRP) policy which evicts data based on both re-reference prediction value and the access recency information. To make the replacement policy more adaptive across different workloads and different phases during execution, Dynamic AI-RRP (DAI-RRP) is proposed which adjusts the insertion position and prediction value for different access patterns. Simulation results show DAI-RRP reduces CPI over LRU and Dynamic RRIP by an average of 8.3% and 4.1% respectively on a single-core processor with a 1MB 16-way set last-level cache (LLC). Evaluations on quad-core CMP with a 4MB shared LLC show that DAI-RRP outperforms LRU and Dynamic RRIP (DRRIP) on the weighted speedup metric by an average of 13.2% and 26.7% respectively. Furthermore, compred to LRU, DAI-RRP requires similar hardware, or even less hardware for high-associativity cache.