Vehicular Digital Twins (VDTs) are emerging as key components in Autonomous and Connected Vehicles (ACVs), enabling real-time monitoring and threat detection. However, in decentralized environments, collaborative learning among VDTs raises significant privacy concerns such as data leakage, re-identification risks and compromised VDTs. This paper presents a decentralized Federated Learning (FL) framework with Differential Privacy (DP) to enable secure and privacy-preserving threat detection across VDTs using a Bi-directional Long Short-Term Memory Autoencoder (Bi-LSTM AE) within an FL setting. The model is designed to capture temporal dependencies in a time series vehicular dataset without compromising data privacy. The model’s performance was evaluated on two attack scenarios, Denial of Service (DoS) and data replay, and compared against existing LSTM-based models in both federated and non-federated contexts. Without DP, the model achieved an F1-score of about 90% and AUC of 95%, performing better than similar methods. We incorporate DP into the FL process to further enhance privacy and assess its impact across different clients. Although, recall remains high, there is a trade-off in precision and accuracy resulting in a global F1-score of 77%. These results highlight the effectiveness of Bi-LSTM AE for decentralized anomaly detection as well as the critical need to balance privacy with utility in real-world deployments.

Lung cancer remains the leading cause of cancer-related deaths worldwide, mainly due to late diagnosis and limited availability of expert pathologists. Although Artificial Intelligence (AI) and DT technologies offer promising avenues for early detection, their adoption in clinical settings introduces significant concerns around data security, system vulnerabilities, and model trust. In response, this paper proposes a novel, trusted medical AI framework that combines blockchain-based device authentication, explainable artificial intelligence (XAI), and DT technologies to enhance diagnostic accuracy, data integrity, and system resilience. The proposed system leverages ResNet for lung condition classification from CT scans, augmented by Grad-CAM for visual explainability, enabling clinicians to interpret AI-driven decisions confidently. A blockchain-based whitelist mechanism authenticates medical devices before data contribution, mitigating risks of tampered or unverified input. Furthermore, the framework integrates explainable digital twin visualization to simulate patient-specific predictions and embeds a vulnerability detection layer to identify and mitigate software flaws in medical IoT and DT components. This comprehensive solution addresses key challenges in real-world healthcare: ensuring data authenticity, interpretability, and cyber resilience, paving the way for secure, transparent, and trustworthy AI-driven diagnostics.

The increasing prevalence of cancer necessitates advanced methodologies for early detection and diagnosis. Early intervention is crucial for improving patient outcomes and reducing the overall burden on healthcare systems. Traditional centralized methods of medical image analysis pose significant risks to patient privacy and data security, as they require the aggregation of sensitive information in a single location. Furthermore, these methods often suffer from limitations related to data diversity and scalability, hindering the development of universally robust diagnostic models. Recent advancements in machine learning, particularly deep learning, have shown promise in enhancing medical image analysis. However, the need to access large and diverse datasets for training these models introduces challenges in maintaining patient confidentiality and adhering to strict data protection regulations. This paper introduces FedViTBloc, a secure and privacy-enhanced framework for medical image analysis utilizing Federated Learning (FL) combined with Vision Transformers (ViT) and blockchain technology. The proposed system ensures patient data privacy and security through fully homomorphic encryption and differential privacy techniques. By employing a decentralized FL approach, multiple medical institutions can collaboratively train a robust deep-learning model without sharing raw data. Blockchain integration further enhances the security and trustworthiness of the FL process by managing client registration and ensuring secure onboarding of participants. Experimental results demonstrate the effectiveness of FedViTBloc in medical image analysis while maintaining stringent privacy standards, achieving 67% accuracy and reducing loss below 2 across 10 clients, ensuring scalability and robustness.

To address the issue of nested entities in named entity recognition, this paper introduces a novel span classification model called SpCL. Unlike traditional models that solely consider the boundary information of a span, SpCL captures the internal features of the span by integrating features at multiple levels, including character-level, word-level, and context-level, in addition to the span’s width feature. Furthermore, SpCL captures external features of the span, especially its "similar entities", which have the same type or boundary words. SpCL uses contrastive learning to bring similar entities closer to each other, connects "similar entities" to construct a span graph, and employs a graph convolutional network to extract feature of the span sub-graph to enhance the current span representation. SpCL leverages the combined information of spans and corresponding "similar entities" to improve the performance of span classification model. Experimental results show that SpCL achieves a significant performance improvement, showing enhanced precision, recall, and F1 scores in comparison to other graph-based models and classical span classification models when applied to datasets such as ACE2004, ACE2005, and GENIA.

In recent years, immense developments have occurred in the field of Artificial Intelligence (AI) and the spread of broadband and ubiquitous connectivity technologies. This has led to the development and commercialization of Digital Twin (DT) technology. The widespread adoption of DT has resulted in a new network paradigm called Digital Twin Networks (DTNs), which orchestrate through the networks of ubiquitous DTs and their corresponding physical assets. DTNs create virtual twins of physical objects via DT technology and realize the co-evolution between physical and virtual spaces through data processing, computing, and DT modeling. The high volume of user data and the ubiquitous communication systems in DTNs come with their own set of challenges. The most serious issue here is with respect to user data privacy and security because users of most applications are unaware of the data that they are sharing with these platforms and are naive in understanding the implications of the data breaches. Also, currently, there is not enough literature that focuses on privacy and security issues in DTN applications. In this survey, we first provide a clear idea of the components of DTNs and the common metrics used in literature to assess their performance. Next, we offer a standard network model that applies to most DTN applications to provide a better understanding of DTN’s complex and interleaved communications and the respective components. We then shed light on the common applications where DTNs have been adapted heavily and the privacy and security issues arising from the DTNs. We also provide different privacy and security countermeasures to address the previously mentioned issues in DTNs and list some state-of-the-art tools to mitigate the issues. Finally, we provide some open research issues and problems in the field of DTN privacy and security.

Collaborative Mobile Crowdsourcing (CMCS) allows platforms to recruit worker teams to collaboratively execute complex sensing tasks. The efficiency of such collaborations could be influenced by trust relationships among workers. To obtain the asymmetric trust values among all workers in the social network, the Trust Reinforcement Evaluation Framework (TREF) based on Graph Convolutional Neural Networks (GCNs) is proposed in this paper. The task completion effect is comprehensively calculated by considering the workers' ability benefits, distance benefits, and trust benefits in this paper. The worker recruitment problem is modeled as an Undirected Complete Recruitment Graph (UCRG), for which a specific Tabu Search Recruitment (TSR) algorithm solution is proposed. An optimal execution team is recruited for each task by the TSR algorithm, and the collaboration team for the task is obtained under the constraint of privacy loss. To enhance the efficiency of the recruitment algorithm on a large scale and scope, the Mini-Batch K-Means clustering algorithm and edge computing technology are introduced, enabling distributed worker recruitment. Lastly, extensive experiments conducted on five real datasets validate that the recruitment algorithm proposed in this paper outperforms other baselines. Additionally, TREF proposed herein surpasses the performance of state-of-the-art trust evaluation methods in the literature.

Cross-modal image-recipe retrieval aims to capture the correlation between food images and recipes. While existing methods have demonstrated good performance on retrieval tasks, they often overlook two crucial aspects: (1) the capture of fine-grained recipe information and (2) the consideration of correlations between embeddings from different modalities. We introduce the Multimodal Fusion Retrieval Framework (MFRF) to address these issues. The proposed framework utilizes a deep learning-based encoder to process recipe and image data effectively, incorporates a fusion network to learn cross-modal semantic alignment, and ultimately achieves image-recipe retrieval. MFRF comprises three integral modules. The recipe preprocessing module utilizes various levels of Transformer to extract essential features such as the title and ingredients from the recipe. Additionally, it employs LSTM based on BERT to establish contextual relationships and dependencies among sentences in the recipe instructions. The multimodal fusion module incorporates visual-linguistic contrastive losses to align the representations of both images and recipes. Moreover, it leverages cross-modal attention mechanisms to facilitate effective interaction between the two modalities. Lastly, the cross-modal retrieval module employs a triple loss function to enable cross-modal retrieval of image-recipe pairs. Experimental evaluations conducted on the widely-used Recipe1M benchmark dataset demonstrate the effectiveness of the proposed MFRF, achieving substantial performance improvements on both the 1k and 10k test sets. Specifically, the results indicate an increase of +9.9% (64.8 R@1) and +8.4% (33.7 R@1) respectively.

With the advent of intelligent technology, the users of spatio-temporal crowdsourcing and their participation in the crowdsourcing tasks continue to increase exponentially. This poses new challenges to the crowdsourcing field. One of the core research areas of spatio-temporal crowdsourcing is task assignment. Most of the existing research on task assignment is focused on offline optimal task assignment, where, the platform has already learned all the information about workers and tasks beforehand. However, these studies cannot obtain good results in real-world situations. At the same time, online task assignment problems often result in local optimal assignment. To solve these problems, more attention needs to be paid to online task assignments and the arrival time of workers. This paper proposes an Online Bilateral Assignment (OBA) problem based on the online assignment model. The competitive ratio of the Greedy algorithm is analyzed according to the OBA problem model. Also, another solution to the OBA problem according to the Greedy algorithm, the Improved-Baseline algorithm, is proposed. Additionally, a Bilateral Online Priority Reassignment algorithm (BOPR) is proposed. The BOPR algorithm realizes real-time task/worker assignment through the bilateral assignment as a solution for online task assignment. In order to guarantee the number of matching tasks, a priority queue is designed in the BOPR algorithm. Considering the waiting time deadlines of tasks and workers and the error rate for priority ranking, it avoids tasks and workers waiting too long and assigns each task to the best possible extent. On this basis, a two-stage assignment strategy is designed for unsuccessful tasks, which could minimize the error rate of the task and significantly improve the efficiency of task assignment. Finally, through experiments on real data sets, the algorithm's performance in terms of global utility value and the number of matches is evaluated.