This paper investigates backdoor attacks in image-oriented semantic communications. The threat of backdoor attacks on symbol reconstruction in semantic communication (SemCom) systems has received limited attention. Previous research on backdoor attacks targeting SemCom symbol reconstruction primarily focuses on input-level triggers, which are impractical in scenarios with strict input constraints. In this paper, we propose a novel channel-triggered backdoor attack (CT-BA) framework that exploits inherent wireless channel characteristics as activation triggers. Our key innovation involves utilizing fundamental channel statistics parameters, specifically channel gain with different fading distributions or channel noise with different power, as potential triggers. This approach enhances stealth by eliminating explicit input manipulation, provides flexibility through trigger selection from diverse channel conditions, and enables automatic activation via natural channel variations without adversary intervention. We extensively evaluate CT-BA across four joint source-channel coding (JSCC) communication system architectures and three benchmark datasets. Simulation results demonstrate that our attack achieves near-perfect attack success rate (ASR) while maintaining effective stealth. Finally, we discuss potential defense mechanisms against such attacks.

Hyperparameter optimization (HPO) is crucial for federated learning (FL) performance. Given the inherent data heterogeneity across clients, recent research has focused on providing personalized hyperparameters for individual clients. However, such personalized approaches introduce exponential search complexity as the number of clients increases, significantly reducing the efficiency of existing HPO methods. To address this challenge, we propose pFedDHPO, a novel personalized HPO framework that efficiently optimizes hyperparameters in a differentiable manner. Specifically, pFedDHPO formulates personalized HPO as an optimization problem targeting joint distribution parameters within the clients’ search space and leverages gradient information from differentiable validation loss to substantially enhance the efficiency of the HPO process. Experimental results demonstrate that pFedDHPO achieves state-of-the-art performance compared to baseline methods, improving accuracy by up to 18.35% under extreme Non-IID data distributions. Additionally, the framework reduces communication overhead by 41.2% compared to conventional HPO methods, making it highly scalable for resource-constrained FL deployments.

Federated Graph Learning (FGL) has emerged as a powerful paradigm for decentralized training of graph neural networks while preserving data privacy. However, existing FGL methods are predominantly designed for static graphs and rely on parameter averaging or distribution alignment, which implicitly assume that all features are equally transferable across clients, overlooking both the spatial and temporal heterogeneity and the presence of client-specific knowledge in real-world graphs. In this work, we identify that such assumptions create a vicious cycle of spurious representation entanglement, client-specific interference, and negative transfer, degrading generalization performance in Federated Learning over Dynamic Spatio-Temporal Graphs (FSTG). To address this issue, we propose a novel causality-inspired framework named SC-FSGL, which explicitly decouples transferable causal knowledge from client-specific noise through representation-level interventions. Specifically, we introduce a Conditional Separation Module that simulates soft interventions through client conditioned masks, enabling the disentanglement of invariant spatio-temporal causal factors from spurious signals and mitigating representation entanglement caused by client heterogeneity. In addition, we propose a Causal Codebook that clusters causal prototypes and aligns local representations via contrastive learning, promoting cross-client consistency and facilitating knowledge sharing across diverse spatio-temporal patterns. Experiments on five diverse heterogeneity Spatio-Temporal Graph (STG) datasets show that SC-FSGL outperforms state-of-the-art methods.

Federated Learning (FL) has emerged as a pivotal paradigm for multi-client collaborative learning, primarily due to its inherent capability to safeguard privacy. Nonetheless, the heterogeneity among FL clients, characterized by their disparate resource capabilities and not independent and identical (NonIID) local datasets, presents a significant challenge. Specifically, low-resource clients, such as edge devices, grapple with the inadequacy to accommodate the entire model parameter set for training purposes. To mitigate this issue, preceding research has ventured into devising methodologies that entail extracting sub-models from the overarching global model, tailored to the specific communication, computational, and memory constraints of individual clients. Despite these advancements, prevailing sub-model extraction techniques, which predominantly hinge on pre-established rules, overlook a crucial factor: the impact of NonIID local data on the trajectory of neuron update dynamics. This oversight can amplify discrepancies between the practical local updates inferred by the sub-model and those anticipated via the entire model, thereby undermining overall performance. In this paper, we introduce FedGSE, an innovative Gradient-based Neuron Selection methodology designed explicitly for FL environments. This methodology aims to curate sub-models that significantly reduce discrepancies in local updates, enhancing alignment with the global model's learning trajectory. Central to the FedGSE approach is a sophisticated algorithm that handpicks critical neurons for sub-model construction. These neurons are identified through their pronounced gradient magnitudes, resulting from the training of the global model on a dataset mirroring the client's data distribution. Consequently, the sub-model's induced local gradient updates closely emulate those derived from directly training the client's data on the full global model, fostering enhanced alignment and performance. Extensive experiments over diverse datasets and tasks demonstrate the superiority of FedGSE over existing baselines.

Federated graph learning excels in learning graph-structured data that are distributed across multiple clients. However, the partition of graph data results in each client only possessing a subgraph, lacking its neighbor nodes, which significantly degrades accuracy. Although exchanging original nodes can address this issue, it requires interaction with a remote server, not only causing significant communication delays but also leaking data privacy. To tackle this, this paper proposes an edge-server-assisted federated graph learning approach, namely FedEGL, which aggregates and exchanges intermediate features of approximated nodes through a third-party edge server, performing cross-client feature alignment and dynamic weighted aggregation while dynamically allocating privacy budgets with adaptive differential privacy to preserve node privacy. Additionally, differential privacy is introduced to protect the privacy of approximated node features by dynamically allocating privacy budgets. Experimental results show that our method achieves accuracy close to that in centralized settings, with the classification accuracy improved by up to 8% compared to the latest baseline. This method can improve model accuracy while protecting privacy, providing an effective solution to the subgraph partitioning problem in federated graph learning.

Foundation model (FM) powered agent services are regarded as a promising solution to develop intelligent and personalized applications for advancing toward Artificial General Intelligence (AGI). To achieve high reliability and scalability in deploying these agent services, it is essential to collaboratively optimize computational and communication resources, thereby ensuring effective resource allocation and seamless service delivery. In pursuit of this vision, this paper proposes a unified framework aimed at providing a comprehensive survey on deploying FM-based agent services across heterogeneous devices, with the emphasis on the integration of model and resource optimization to establish a robust infrastructure for these services. Particularly, this paper begins with exploring various low-level optimization strategies during inference and studies approaches that enhance system scalability, such as parallelism techniques and resource scaling methods. The paper then discusses several prominent FMs and investigates research efforts focused on inference acceleration, including techniques such as model compression and token reduction. Moreover, the paper also investigates critical components for constructing agent services and highlights notable intelligent applications. Finally, the paper presents potential research directions for developing real-time agent services with high Quality of Service (QoS).

Tor is a low-latency anonymous communication network that protects user privacy by encrypting website traffic. However, recent website fingerprinting (WF) attacks have shown that encrypted traffic can still leak users' visited websites by exploiting statistical features such as packet size, direction, and inter-arrival time. Most existing WF attacks formulate the problem as a single-tab classification task, which significantly limits their effectiveness in realistic browsing scenarios where users access multiple websites concurrently, resulting in mixed traffic traces. To this end, we propose PrismWF, a multi-granularity patch-based Transformer for multi-tab WF attack. Specifically, we design a robust traffic feature representation for raw web traffic traces and extract multi-granularity features using convolutional kernels with different receptive fields. To effectively integrate information across temporal scales, the proposed model refines features through three hierarchical interaction mechanisms: inter-granularity detail supplementation from fine to coarse granularities, intra-granularity patch interaction with dedicated router tokens, and router-guided dual-level intra- and cross-granularity fusion. This design aligns with the cognitive logic of global coarse-grained reconnaissance and local fine-grained querying, enabling effective modeling of mixed traffic patterns in WF attack scenarios. Extensive experiments on various datasets and WF defenses demonstrate that our method achieves state-of-the-art performance compared to existing baselines.

Serving Mixture-of-Experts (MoE) large language models across distributed edge servers is bottlenecked by the cross-server expert transmission. The existing approaches mainly focus on how to reach a remote expert faster. However, in this paper, we instead consider whether a given layer, and the layers after it, need to be executed at all. To this end, a communication-aware adaptive-depth framework is proposed in this paper, termed TrimMoE, which couples layer skipping and confidence-based early exit with substitute execution and server-expert selection under a unified quality budget. Specifically, in the offline stage, TrimMoE freezes the backbone, trains the lightweight per-layer exit heads, calibrates the per-layer importance thresholds, and allocates the expert replicas by a skip/exit-aware redundancy benefit. In the online stage, a transition-aware look-ahead anticipates the token movement, so that the depth reduction targets the costliest transmissions, and besides, two feedback rules adapt the delay-quality weights and the exit threshold. Moreover, we prove that the substitution-and-skipping proxy degradation never exceeds the configured budget, and that the early exit is admitted only under a calibrated confidence gate. On a heterogeneous 10-server testbed with Switch-Base-8E, Qwen-MoE-A2.7B, and Mixtral-8x7B, TrimMoE reduces the average latency by up to 62.8

The rapid growth of connected vehicles creates new opportunities to exploit internet of vehicles (IoV) data for traffic state estimation (TSE), which is a key enabler of intelligent transportation systems (ITS). In this paper, we propose a cost-effective TSE framework that leverages sparse IoV data, which significantly reduces the data collection overhead associated with large-scale IoV datasets. We further analyze the impact of data sparsification and show that the induced estimation errors can be well approximated by Gaussian noise, thereby reformulating sparse IoV-based TSE as a denoising problem. To enhance estimation accuracy, we develop a spatial-temporal attention model, termed the convolutional retentive network (CRNet), which integrates convolutional neural networks (CNNs) for spatial correlation learning with a retentive network (RetNet) for temporal dependency modeling. Extensive experiments conducted on a large-scale real-world IoV dataset validate the feasibility of TSE under sparse IoV sensing conditions. Notably, even when only 5% of the data is available, CRNet achieves a mean absolute error (MAE) below 5 km/h, demonstrating both the high accuracy of the proposed approach and its practical applicability in real-world scenarios.

Federated Graph Learning is an efficient technique for processing graph data, capable of integrating information from various data sources in a distributed environment. However, graph data from different domains often exhibit significant differences in both features and structures, which is known as the heterogeneity problem of cross-domain federated graph data. Existing methods still have limitations in the underlying extraction and aggregation of structural information for cross-domain federated graphs. To address this issue, we propose FedCHG: Graph Autoencoder Enhanced Federated Learning for Cross-Domain Heterogeneous Graph. Specifically, we combine the Graphlet subgraph algorithm and the random walk diffusion algorithm to extract graph structural information from both local and global perspectives, constructing a universal structural representation. We design a graph encoder-decoder architecture to integrate global structural information while preserving local structural information. Additionally, we introduce a weighted aggregation strategy based on graph structural similarity, which considers structural differences during the aggregation process and enhances the overall performance of the model. Experiments on publicly available multi-domain graph datasets show that compared to current state-of-the-art federated graph learning baselines, the proposed FedCHG improves the accuracy by up to 5% in cross-domain scenarios, demonstrating the effectiveness of FedCHG in collaborative learning scenarios in various graph datasets.

The increase of bandwidth-intensive applications in sixth-generation (6G) wireless networks, such as real-time volumetric streaming and multi-sensory extended reality, demands intelligent multicast routing solutions capable of delivering differentiated quality-of-service (QoS) at scale. Traditional shortest-path and multicast routing algorithms are either computationally prohibitive or structurally rigid, and they often fail to support heterogeneous user demands, leading to suboptimal resource utilization. Neural network-based approaches, while offering improved inference speed, typically lack topological generalization and scalability. To address these limitations, this paper presents a graph neural network (GNN)-based multicast routing framework that jointly minimizes total transmission cost and supports user-specific video quality requirements. The routing problem is formulated as a constrained minimum-flow optimization task, and a reinforcement learning algorithm is developed to sequentially construct efficient multicast trees by reusing paths and adapting to network dynamics. A graph attention network (GAT) is employed as the encoder to extract context-aware node embeddings, while a long short-term memory (LSTM) module models the sequential dependencies in routing decisions. Extensive simulations demonstrate that the proposed method closely approximates optimal dynamic programming-based solutions while significantly reducing computational complexity. The results also confirm strong generalization to large-scale and dynamic network topologies, highlighting the method's potential for real-time deployment in 6G multimedia delivery scenarios. Code is available at https://github.com/UNIC-Lab/GNN-Routing.

The deployment of large language models' (LLMs) inference at the edge can facilitate prompt service responsiveness while protecting user privacy. However, it is critically challenged by the resource constraints of a single edge node. Distributed inference has emerged to aggregate and leverage computational resources across multiple devices. Yet, existing methods typically require strict synchronization, which is often infeasible due to the unreliable network conditions. In this paper, we propose HALO, a novel framework that can boost the distributed LLM inference in lossy edge network. The core idea is to enable a relaxed yet effective synchronization by strategically allocating less critical neuron groups to unstable devices, thus avoiding the excessive waiting time incurred by delayed packets. HALO introduces three key mechanisms: (1) a semantic-aware predictor to assess the significance of neuron groups prior to activation. (2) a parallel execution scheme of neuron group loading during the model inference. (3) a load-balancing scheduler that efficiently orchestrates multiple devices with heterogeneous resources. Experimental results from a Raspberry Pi cluster demonstrate that HALO achieves a 3.41x end-to-end speedup for LLaMA-series LLMs under unreliable network conditions. It maintains performance comparable to optimal conditions and significantly outperforms the state-of-the-art in various scenarios.

Although mixture-of-experts, MoE, models have been increasingly adopted to scale large language models with moderate computation cost, it remains challenging to deploy MoE inference over resource-constrained and bandwidth-limited edge infrastructures. Existing distributed MoE serving methods mainly rely on exact expert placement, caching, replication, or communication scheduling, while overlooking the functional similarity among experts, which provides an opportunity to reduce cross-server token transmission. Therefore, this paper introduces a similarity-aware expert allocation and distributed deployment framework, dubbed OrderMoE, which aims to accelerate edge MoE inference while balancing inference latency, communication overhead, server workload, and inference quality. OrderMoE first constructs an expert similarity model based on router-induced logits representations and partitions experts in each MoE layer into multiple similarity groups. Then, it develops a similarity-aware expert grouping and deployment strategy to improve local similarity coverage across edge servers. Since reducing remote expert invocation and preserving exact inference quality are conflicting objectives, OrderMoE further designs a quality-aware and trajectory-aware runtime server-expert selection algorithm to decide whether a token should invoke its remote target expert or use a feasible local substitute expert. Experimental results on a real distributed edge testbed show that OrderMoE significantly reduces average latency, tail latency, cross-server traffic, and remote expert invocation ratio, while introducing only small and controllable inference quality degradation.

Non-Centralized Continual Learning (NCCL) has become an emerging paradigm for enabling distributed devices such as vehicles and servers to handle streaming data from a joint non-stationary environment. To achieve high reliability and scalability in deploying this paradigm in distributed systems, it is essential to overcome challenges stemming from both spatial and temporal dimensions, manifesting as distribution shifts, catastrophic forgetting, heterogeneity, and privacy issues. This survey focuses on a comprehensive examination of the development of the non-centralized continual learning algorithms and the real-world deployment across distributed devices. We begin with an introduction to the background and fundamentals of non-centralized learning and continual learning. Then, we review existing solutions from three levels to represent how existing techniques alleviate the catastrophic forgetting and distribution shift. Additionally, we delve into the various types of heterogeneity issues, security, and privacy attributes, as well as real-world applications across three prevalent scenarios. Furthermore, we establish a large-scale benchmark to revisit this problem and analyze the performance of the state-of-the-art NCCL approaches. Finally, we discuss the important challenges and future research directions in NCCL.

Radio maps (RMs) provide the digital representation of the wireless propagation environment, mapping complex geographical and topological boundary conditions to critical spatial-spectral metrics that range from received signal strength to full channel state information matrices. The integration of artificial intelligence into next generation wireless networks further necessitates the accurate construction of RMs as a foundational prerequisite for electromagnetic digital twins. This paper presents a comprehensive survey of learning-based RM construction, systematically addressing three intertwined dimensions: data, paradigms, and physics-awareness. From the data perspective, we review physical measurement campaigns, ray tracing simulation engines, and publicly available benchmark datasets, identifying their respective strengths and fundamental limitations. From the paradigm perspective, we establish a core taxonomy that categorizes RM construction into source-aware forward prediction and source agnostic inverse reconstruction, and examine five principal neural architecture families spanning convolutional neural networks, vision transformers, graph neural networks, generative adversarial networks, and diffusion models. We further survey optics-inspired methods adapted from neural radiance fields and 3D Gaussian splatting for continuous wireless radiation field modeling. From the physics-awareness perspective, we introduce a three-level integration framework encompassing data-level feature engineering, loss-level partial differential equation regularization, and architecture level structural isomorphism. Open challenges including foundation model development, physical hallucination detection, and mortized inference for real-time deployment are discussed to outline future research directions. The project page is at https://github.com/UNIC-Lab/Awesome-Radio-Map-Categorized.

Mixture-of-Experts (MoE) models have become a dominant architecture for large-scale AI services, yet deploying them over geo-distributed heterogeneous edge servers remains challenging. When the Top-k activated experts of a token are spread across multiple servers, the optimal routing depends jointly on cross-server link bandwidth, heterogeneous GPU computing capability, GPU-CPU expert loading delay, instantaneous queueing backlog, and replica-level quantization quality loss. Existing distributed inference and MoE serving methods address these factors separately and do not provide a unified framework for online multi-server collaborative routing. In this paper, we propose HetRoute, a heterogeneous-cost-aware collaborative routing framework for distributed edge MoE inference. HetRoute introduces a unified per-assignment cost model that explicitly captures four cost components: cross-server transmission, GPU-CPU offloading, GPU computation with queueing, and quantization-induced quality penalty. Guided by this model, the offline stage determines expert server placement, GPU-CPU residency, and replica precision through a routing-cost-coupled deployment algorithm, while the online stage routes the Top-k activated expert set as a whole by minimizing the bottleneck layer cost via exact enumeration or beam search. Theoretical analysis establishes fallback feasibility, a bound on the number of participating servers, per-layer optimality for small candidate domains, and online computational complexity. Trace-driven evaluation on three MoE models over a heterogeneous 10-server edge testbed shows that HetRoute reduces average inference latency by up to 59.0