The proliferation of smart vehicles and resource-hungry applications has imposed challenges to on-board systems. By exploiting the clustered vehicles in neighbor-following network (NFN), the multi-hop vehicular network (MHVN) enables cooperative communication among multiple vehicles, making it promising for vehicular edge computing (VEC). Nonetheless, to the best of our knowledge, the joint task offloading and resource allocation strategies in MHVN remains open due to challenges arising from the prevalence of link disruptions, the substantial variability in channel states, and the limited computational resource. With the above considerations, in this work, the task offloading and resource allocation are jointly optimized in the MHVN to minimize the offloading cost and maximize the task completion rate (TCR). However, the optimization problem turns out to be a mixed integer nonlinear programming (MINLP) problem. To this end, the problem is decoupled into two subproblems, which are solved by graph theory and improved water-filling algorithm, respectively. Furthermore, an iterative optimization algorithm is designed to mitigate the impact of the relaxation of delay constraints. Simulations are conducted to confirm the effectiveness of the proposed scheme.

To enable large model (LM) based edge intelligent service provisioning, on-device fine-tuning with locally personalized data allows for continuous and privacy-preserving LM customization. In this paper, we propose RingAda, a collaborative training framework designed for fine-tuning transformer-based LMs on edge devices. Particularly, RingAda performs parameterefficient adapter fine-tuning across a set of interconnected edge devices, forming a ring topology for per-batch training by sequentially placing frozen transformer blocks and their trainable adapter modules on the devices. RingAda follows a novel pipeline-parallel training mechanism with top-down adapter unfreezing, allowing for early-stopping of backpropagation at the lowest unfrozen adapter layer, thereby accelerating the finetuning process. Extensive experimental results demonstrate that RingAda significantly reduces fine-tuning time and memory costs while maintaining competitive model performance compared to its peer designs.

This paper introduces a novel multimodal emotion recognition model, the Modality-Adaptive Masked Fusion Network (MAMF-Net), designed to mitigate information loss and improve cross-modal alignment during the fusion of speech and text modalities. MAMF-Net employs an audio-guided text encoder to enhance the semantic representation of text by leveraging the temporal resolution and contextual information inherent in speech, thereby ensuring accurate alignment of modal features. Additionally, the model utilizes a modality transfer-based MAE masking strategy, which effectively captures complementary information between modalities by partially masking transferred information, thus improving fusion effectiveness and system stability. The experimental results show that MAMF-Net outperforms existing methods on datasets such as CMU-MOSI and CMU-MOSEI, highlighting its significant potential for multimodal emotion analysis.

The proliferation of Internet of Things (IoT)-generated distributed personal data enables user-specific large language model (LLM) adaptation at the edge. The split federated learning (SFL) facilitates collaborative learning and reduces memory footprint by model splitting, which necessitates the transmission of intermediate activations, rendering it susceptible to reconstruction attacks and privacy breaches. In this paper, we present a privacy-aware SFL scheme addressing the accuracy-efficiency-privacy trilemma in LLM fine-tuning over heterogeneous IoT devices. Particularly, we develop a privacy quantification metric based on Fisher information to assess layer-wise privacy risks in smashed data transmission. Guided by this metric, we establish an analytical model that captures the intricate relationships between privacy leakage, fine-tuning convergence time, and device energy consumption. To optimize these three aspects, we formulate a multi-objective mixed-integer programming problem. Then, an -constraint-based block coordinate descent (BCD) algorithm is proposed to jointly determine the optimal LLM split layer, transmit power, and bandwidth allocation for IoT devices under their memory and network constraints. Extensive simulation results demonstrate the proposed scheme’s effectiveness in achieving 24% faster convergence, 40% lower energy consumption, and 7% reduced privacy leakage compared to baseline approaches, while maintaining competitive model accuracy.

Vehicular fog computing (VFC) that supports inter-vehicular task offloading emerges as a promising complement to handle the explosive growth of computation-intensive tasks in Intelligent Transportation Systems (ITS). Nonetheless, as the fog access points (F-APs) in crowed areas are often overloaded, the conventional single F-AP VFC may become incompetent and energy-inefficient. To tackle the issue, a novel scheme of non-orthogonal multiple access (NOMA)-enabled multi-F-AP VFC with partial offloading is proposed in this work. However, the corresponding energy minimization turns out to be a highly non-trivial non-linear mixed-integer programming problem. To this end, the optimal power allocation is derived by exploiting monotonicity while good task splitting ratio and user association are found through successive convex approximation (SCA)-based interior-point method and game theoretic approach, respectively. Extensive simulations based on MATLAB show that, in the considered scenarios, the proposed scheme can fulfill a more balanced offloading and better exploit the available computing resources, thereby leading to an approximately 30% energy consumption reduction compared to the baselines.

Device-to-device (D2D) communication is one of the most promising technologies for relieving the pressure of demands in the 5G mobile networks. However, due to randomness of user request, limitation of storage space and transmission capacity, it is still a challenge for channel allocation to optimize the successful delivery ratio of contents. Thus, in this article, we first consider the link selection problem for mobile users in multi-D2D (MD2D)-based content delivery networks. We establish a content delivery utility that combines physical and social aspects, taking into account energy consumption, user mobility, and trust relationships. Second, we model the content delivery link selection as the maximum weighted matching problem. For this NP-hard problem, by relaxing the integer constraints, we propose a BnB-CDLS algorithm based on the branch-and-bound method for the proposed content delivery scheme. Furthermore, we prove that the problem can be computationally reduced to a monotone submodular problem subject to matroid and knapsack constraints, which can be solved by a greedy GA-CDLS algorithm. Numerical results show that as compared with the existing schemes, the proposed scheme can significantly improve the successful delivery ratio of contents.

Identification of essential proteins based on protein interaction network (PIN) is a very important and hot topic in the post genome era. Up to now, a number of network-based essential protein discovery methods have been proposed. Generally, a static protein interaction network was constructed by using the protein-protein interactions obtained from different experiments or databases. Unfortunately, most of the network-based essential protein discovery methods are sensitive to the reliability of the constructed PIN. In this paper, we propose a new method for constructing refined PIN by using gene expression profiles and subcellular location information. The basic idea behind refining the PIN is that two proteins should have higher possibility to physically interact with each other if they appear together at the same subcellular location and are active together at least at a time point in the cell cycle. The original static PIN is denoted by S-PIN while the final PIN refined by our method is denoted by TS-PIN. To evaluate whether the constructed TS-PIN is more suitable to be used in the identification of essential proteins, 10 network-based essential protein discovery methods (DC, EC, SC, BC, CC, IC, LAC, NC, BN, and DMNC) are applied on it to identify essential proteins. A comparison of TS-PIN and two other networks: S-PIN and NF-APIN (a noise-filtered active PIN constructed by using gene expression data and S-PIN) is implemented on the prediction of essential proteins by using these ten network-based methods. The comparison results show that all of the 10 network-based methods achieve better results when being applied on TS-PIN than that being applied on S-PIN and NF-APIN.