Federated learning is increasingly adopted for traffic flow prediction (TFP) to enable privacy preserving collaboration across distributed sensors. However, real-world deployments are highly heterogeneous in computational capability, causing stragglers that dominate per-round latency and severely slow down model updates. Most existing approaches mitigate stragglers by suppressing or discarding slow clients, which reduce data representativeness and introduce training bias. It is a harmful trade-off for TFP where broad spatial coverage is crucial for accuracy. We propose a communication-efficient logical clustering federated learning framework (LCFed) that mitigates stragglers by logically balancing effective training time while preserving full client participation. LCFed combines a coarse-grained logical dynamic clustering algorithm ( LoDynClust ) to balance computational resources across clusters and reduce synchronization delays, with a fine-grained intra-cluster adaptive collaborative training mechanism ( ICACT ) to regulate aggregation intervals and mitigate training bias. We further provide a convergence analysis. Extensive experiments on three real-world traffic datasets show that LCFed significantly reduces training latency caused by stragglers while maintaining competitive prediction accuracy compared with state-of-the-art baselines.

The growing prevalence of hate videos promoting intolerance, bigotry, and discrimination presents significant psychosocial threats to both individuals and society. Current detection methods often rely on black-box models, which lack interpretability — a crucial factor for fostering more reliable content moderation and trustworthy AI. To bridge this gap, we propose MATCH, the first attempt to achieve interpretable hate video detection via multiple Large Multimodal Model (LMM) agent collaboration. Our method facilitates a noveldiversely generate-then-verifyparadigm, where LMM agents work in tandem to generate diverse clues and verify them to yield more faithful explanations. MATCH begins by proposing a new Dual-Perspective Proposing paradigm, where two LMM agents are regarded asProposersto independently identify evidential clues from opposing angles – hate and non-hate. Leveraging these comprehensive clues, we introduce an innovative Spatiotemporal Evidence-Grounded Verification mechanism. In this mechanism, a third LMM agent acts as aVerifier, rigorously validating and reconciling the proposed clues against spatiotemporal evidences directly extracted from video content, yielding coherent and faithful explanations. Finally, these explanations are integrated with video features, enabling accurate identification of complex and ambiguous hateful content. Extensive experiments conducted on three benchmark datasets demonstrate that MATCH not only achieves state-of-the-art performance, but also provides reliable and trustworthy rationales for the predictions. Code is available at https://anonymous.4open.science/r/MATCH-HVD.

ABSTRACT Trajectory classification techniques are widely applied in personalized recommendation, precision marketing, and public opinion analysis. However, these models are susceptible to adversarial attacks, where subtle and imperceptible perturbations can mislead the models and significantly compromise classification accuracy. Existing adversarial attacks for trajectory classification often adopt black‐box strategies or single‐step gradient updates, directly perturbing raw trajectory data while neglecting the influence of trajectory features on classification. To address this issue, we propose a feature‐level adversarial attack framework that introduces perturbations at the trajectory feature level rather than on raw data. Specifically, trajectory features are first extracted using the spatial encoding‐based extraction algorithm. The MMOE model is then employed to compute feature importance weights, based on which minimal perturbations are added to the most influential features to maximize classification disruption while preserving trajectory structure. In addition, we develop a differential privacy‐enhanced defense strategy, which adds optimal Laplace noise to the trajectory features and performs inverse transformation to mitigate the impact of adversarial perturbations. Experiments on real‐world datasets show that the proposed adversarial attack successfully deceives the classification model, while the defense strategy significantly improves its robustness. Experimental code can be found at: https://github.com/whwh456/Trattack .

The growing adoption of 3D point cloud in applications like autonomous driving has heightened concerns about their vulnerability to adversarial attacks. Existing defense methods face two fundamental challenges: ineffective detection of imperceptible adversarial examples and poor restoration of severely distorted point cloud. In this paper, we present ADDR, an end-to-end defense framework that integratesBinary Geometric Feature Anomaly Detection (BGFAD)andDistorted point cloud Restoration (DPCR). BGFAD employs a dual threshold mechanism combining global distance statistics and local curvature analysis to detect both substantial and imperceptible adversarial perturbations. DPCR leverages attention enhanced feature encoding to reconstruct missing geometric structures while preserving semantic integrity through bidirectional Chamfer loss optimization. Our framework uniquely bridges traditional geometric priors with deep learning mechanisms, achieving attack-agnostic defense without classifier retraining. Extensive experiments on ModelNet40, ShapeNet and ScanObjectNN datasets demonstrate state-of-the-art performance, with about 12% higher robustness against structural attacks and 6× better restoration fidelity than existing methods. ADDR maintains real-time processing capabilities while reducing adversarial success rates to <5% across diverse attacks. The code is available at https://github.com/whwh456/ADDR.

Federated learning (FL) can complete collaborative model training without transferring local data, which can greatly improve the training efficiency. However, FL is susceptible data and model backdoor attacks. To address data backdoor attack, in this article, we propose a defense method named TSF. TSF transforms data from time domain to frequency domain and subsequently designs a low-pass filter to mitigate the impact of high-frequency signals introduced by backdoor samples. Additionally, we undergo homomorphic encryption on local updates to prevent the server from inferring user’s data. We also introduce a defense method against model backdoor attack named ciphertext field similarity detect differential privacy (CFSD-DP). CFSD-DP screens malicious updates using cosine similarity detection in the ciphertext domain. It perturbs the global model using differential privacy mechanism to mitigate the impact of model backdoor attack. It can effectively detect malicious updates and safeguard the privacy of the global model. Experimental results show that the proposed TSF and CFSD-DP have 73.8% degradation in backdoor accuracy while only 3% impact on the main task accuracy compared with state-of-the-art schemes. Code is available athttps://github.com/whwh456/TSF.

This paper studies the age of information (AoI) and energy tradeoff (AET) problem in an aerial-ground collaborative mobile edge computing system, where a high-altitude platform and an unmanned aerial vehicle (UAV) work together to offer computing services for ground devices (GDs). The AET problem is formulated as a multi-objective optimization problem (MOP) that aims at simultaneously minimizing the total AoI of GDs and total energy consumption of the UAV by optimizing its flight paths and task offloading ratios. Addressing the AET problem poses a significant challenge due to the inherent conflict between the two objectives. The existing methods cannot well address the MOP because they adopt the linear combination to transform an MOP into a single-objective optimization problem using fixed weights (i.e., preferences), ignoring the conflict between objectives. Moreover, user preferences may change over time in dynamic MEC systems. To overcome these challenges, we first build a multi-objective Markov decision process model with a vectorial reward for the AET problem. There are one-to-one relationships between each component of the reward and one of the two objectives. Then, we propose a multi-objective learning algorithm based on proximal policy optimization (PPO), which primarily comprises a training phase and an evolutionary phase. The former adopts multi-objective PPO to iteratively optimize multiple learning individuals, aiming to obtain a nondominated policy set. The latter employs a genetic operator to further improve the quality of each policy in the set. Specifically, the crossover and mutation operators operate at the parameter level of policy networks, avoiding stagnation and premature convergence. The experiment results validate that the proposed approach obtains a set of excellent nondominated policies and a favorable balance between objectives. Moreover, the proposed approach achieves improvements of at least 39.8%, 2.1%, and 15.3% regarding AoI, energy consumption, and cost compared with several algorithms.

Abstract Federated learning (FL) can learn a shared global model across multiple client devices without breaching privacy requirements. But an essential challenge is that devices in FL usually have limited network bandwidth, resulting in inefficient communication as an important bottleneck for FL implementation. Current studies try to overcome this shortcoming by compressing the number of model update bits uploaded by every client. But they did not explore the underlying reason why redundant parameters are generated. In this paper, we propose Corset-Based Federated Learning framework (CBFL) —a novel FL communication framework from the perspective of redundancy data. Instead of training full datasets on a regular network model, CBFL trains a much smaller evolution network model on extracted corset, which intrinsically reduces the overall transmission bits and obtains efficient computation while maintaining a desirable model accuracy. In CBFL, a novel Fedcorset construction algorithm at selected clients and a further distributed model evolution scheme to fit the constructed corset are included. The training model size is dynamically adapted to the corset, either removing a fraction of unimportant or adding important connections at each communication iteration. Experimental results show that CBFL transfers about 13% of communication bits and saves around 56% computing time while having only 2% destination in model accuracy.

Abstract Federated learning (FL) is an emerging privacy-preserving technology for machine learning, which enables end devices to cooperatively train a global model without uploading their local sensitive data. Because of limited network bandwidth and considerable communication overhead, communication efficiency has become an essential bottleneck for FL. Existing solutions attempt to improve this situation by reducing communication rounds while usually come with more computation resource consumption or model accuracy deterioration. In this paper, we propose a parameter Prediction-Based DL (PBFL). In which an extended Kalman filter-based prediction algorithm, a practical prediction error threshold setting mechanism and an effective global model updating strategy are included. Instead of collecting all updates from participants, PBFL takes advantage of predicting values to aggregate the model, which substantially reduces required communication rounds while guaranteeing model accuracy. Inspired by the idea of prediction, each participant checks whether its prediction value is out of the tolerance threshold limits and only uploads local updates that have an inaccurate prediction value. In this way, no additional local computational resources are required. Experimental results on both multilayer perceptrons and convolutional neural networks show that PBFL outperforms the state-of-the-art methods and improves the communication efficiency by >66% with 1% higher model accuracy.

Abstract Although analyzing and mining user’s trajectory data can provide outstanding benefit, data owners may not be willing to upload their trajectory data because of privacy concerns. Recently, differential privacy technology has achieved a good trade-off between data utility and privacy preserving by publishing noisy outputs, and relevant schemes have been proposed for trajectory release. However, we experimentally find that a relatively accurate estimate of the true data value can still be obtained from the noisy outputs by means of a posterior estimation. But there are no practical mechanisms against current schemes to verify their effectiveness and resistance. To fill this gap, we propose a solution to evaluate the resistance performance of differential privacy on trajectory data release, including a notion of correlation-distinguishability filtering (CDF) and a privacy quantification measurement. Specifically, taking advantage of the principle of filtering that independent noise can be filtered out from correlated sequence, CDF is proposed to sanitize the noise added into the trajectory. To conduct this notion in practice, we attempt to apply a Kalman/particle filter to filter out the corresponding Gaussian/Laplace noise added by differential privacy schemes. Furthermore, to quantify the distortion of privacy strength before and after filtering, an entropy-based privacy quantification metric is proposed, which is used to measure the lost uncertainty of the true locations for an adversary. Experimental results show that the resistance performance of current approaches has a degradation to varying degrees under the filtering attack model in our solution. Moreover, the privacy quantification metric can be regarded as a unified criterion to measure the privacy strength introduced by the noise that does not conform to the form required by differential privacy.