Federated Learning (FL) enables privacy-preserving distributed training but remains vulnerable to backdoor attacks. Attackers can embed malicious trigger-label associations into the global model by participating in the aggregation process. Existing defense methods typically defend against backdoor attacks by detecting and filtering malicious updates that deviate from benign ones. However, we find that these defenses fail under domain skew, where differing feature distributions across clients increase update heterogeneity, making it harder to distinguish malicious updates from benign ones. To address this challenge, we propose DoBlock, a novel defense that utilizes an aggregatable domain infuser incapable of embedding malicious associations, through federated training to facilitate cross-domain knowledge sharing. Moreover, DoBlock prevents malicious association propagation by isolating local models from aggregation, as local models remain client-specific and rely solely on local data for training. Experiments on five domain skew datasets (Digits, PACS, VLCS, Office-Caltech10, and DomainNet) show that DoBlock maintains attack success rates below 2.5%, while achieving the highest main task accuracy, demonstrating superior robustness without sacrificing benign performance.

Federated Noisy Labels Learning (FNLL) allows global model to be jointly trained on multiple clients with varying degrees of noisy labels while preserving privacy, and despite recent research advances, distinguishing between client clean and noisy samples is still tricky since the distribution of labels among clients is always both noisy and class-imbalanced, leading to the poor performance of existing FNLL methods. To address this problem, we propose a novel framework called FedPN, the first framework to utilize Contrastive Language-Image Pre-training (CLIP) for federated noisy labels tasks. Then, to achieve higher performance for the global model, we introduce an attention based Prototype Adapter to identify more plausible local data for local model training, further improving training stability. We validate the effectiveness of FedPN by conducting extensive experiments on benchmark datasets under both Independently and Identically Distributed (IID) and Non-IID data partitions. The experimental results show that FedPN can effectively filter noisy samples from different clients, and compared with the state of-the-art FNLL method, the FedPN achieves at most and at least 8.39% and 0.88% performance improvement in the case of highly heterogeneous noisy labels.

The rapid growth of Internet of Things (IoT) devices has heightened hardware security concerns, particularly with the emergence of hardware Trojans (HTs) as malicious components in integrated circuits. These HTs pose significant threats to information privacy and system performance. The early detection methods rely on golden references and domain knowledge, limiting their implementation in large-scale integrated circuits. Traditional machine learning-based HT detections can effectively identify infected circuits without golden chips. However, locating and evaluating the specific behavior of HTs in vast and complex circuits is challenging. To address these challenges, this paper presents an HT Object Detection (HTOD) problem, which involves two primary tasks: identifying the boundaries and sizes of HTs within circuits and distinguishing between different HT behaviors. We have developed a two-stage HT object detection framework based on a bidirectional jumping knowledge network called HTOD-BGNN to solve the HTOD problem, which enables progressive refinement of localized regions. It incorporates enhanced base-type features to ensure scalability and information retention during modeling. Additionally, the implementation of sample augmentation alleviates the challenges of sample imbalance and scarcity. The proposed method enables more intelligent circuit detection, achieving 100% and 87.5% detection rates for Trojan triggering and payload behaviors, respectively. It demonstrates superior localization performance and effective generalization to unseen circuit scenarios via its region refinement mechanism, achieving F1-scores of 54.01% on TrustHub and 90.04% on TRIT datasets.

Federated learning remains vulnerable to backdoor attacks through malicious parameter updates, with existing defenses limited by homogeneous data assumptions or reliance on gradient anomaly detection. We reveal that FedAvg's critical flaw lies in malicious feature extractor propagation: aggregating poisoned extractors degrades defense accuracy to <70% across five benchmarks, while benign extractors with poisoned headers retain an average of 89.36% defense accuracy. Therefore, we propose FeatShield, a feature-space isolation framework that prevents backdoor propagation via non-aggregated local extractors trained on clean client data. FeatShield introduces 1) variance-aware alignment, adaptively balancing client-specific features and global consistency using local variance metrics, and 2) adversarial feature synthesis, generating non-linear synthetic features via GAN to enhance the global prediction header's generalization on main tasks. Extensive experiments on eight real-world datasets show that FeatShield achieves the best defense performance. For instance, under heterogeneous data (Dirichlet β=0.5) and strong attacks (50% malicious clients), FeatShield achieves 99.26-99.89% defense accuracy and main task accuracy exceeding FedAvg by 1.32-5.70%, demonstrating its superior resistance to backdoor attacks without sacrificing the benign performance.

Next Point-of-Interest (POI) recommendation plays an important role in various Location-Based Social Networks (LBSNs). It main objective is to predict the user's next interested POI based on previous check-in information. Most existing research treats next POI recommendation as a sequence prediction problem, ignoring collaborative signals from other users as well as security. Instead, a Differential Privacy-based Transformer Framework for Next POI Recommendation (LDPTRec) is proposed, which focuses on dynamic privacy protection, category-specific temporal modeling, and privacy-preserving sequence integration. (1) A privacy trajectory flow graph is constructed by using social-aware edge local differential privacy to protect users behavioral and location privacy. (2) A novel temporal category-aware context embedding algorithm is designed to capture diverse temporal patterns of POI categories. (3) A DP-Transformer algorithm with theoretical privacy guarantees, validated by experiments showing 5.6% accuracy gains and 35.17% lower cold-start latency. Ablation studies validate its components effectiveness, and time-cost experiments confirm its enhanced recommendation efficiency. Overall, LDPTRec effectively balances recommendation security and efficiency while improving accuracy.

Synthetic lethality (SL) has shown great promise for the discovery of novel targets in cancer. CRISPR double-knockout (CDKO) technologies can only screen several hundred genes and their combinations, but not genome-wide. Therefore, good SL prediction models are highly needed for genes and gene pairs selection in CDKO experiments. However, lack of scalable SL properties prevents generalizability of SL interactions to out-of-sample data, thereby hindering modeling efforts. In this paper, we recognize that SL connectivity is a scalable and generalizable SL property. We develop a novel two-step multilayer encoder for individual sample-specific SL prediction model (MLEC-iSL), which predicts SL connectivity first and SL interactions subsequently. MLEC-iSL has three encoders, namely, gene, graph, and transformer encoders. MLEC-iSL achieves high SL prediction performance in K562 (AUPR, 0.73; AUC, 0.72) and Jurkat (AUPR, 0.73; AUC, 0.71) cells, while no existing methods exceed 0.62 AUPR and AUC. The prediction performance of MLEC-iSL is validated in a CDKO experiment in 22Rv1 cells, yielding a 46.8% SL rate among 987 selected gene pairs. The screen also reveals SL dependency between apoptosis and mitosis cell death pathways.

Mobile Ad hoc Networks (MANETs) is a decentralized network where mobile nodes in the network communicate with each other over wireless links without any infrastructure or central controller. With the latest development of unmanned aerial vehicles (UAVs), the development of network technology for multi-cooperative UAVs has become crucial. Routing in a Flying Ad hoc Networks (FANETs) is a challenging problem. However, traditional MANET routing protocols often only consider the current connection state when establishing a route and selecting the next hop. In high mobility FANETs, these methods are possible not optimal or even wrong. Meanwhile, FANET has certain advantages. For example, most UAVs are equipped with Global Position System (GPS) or similar devices that can provide their own spatial location information. Routing decisions can be optimized and adjusted with this available information. In this paper, based on the Optimized Link State Routing (OLSR) protocol, we propose a Link Duration OLSR (LD-OLSR) protocol. The proposed protocol uses the situation information of the nodes in three dimensions to predict the duration of the link to optimize the establishment of the backbone network. The simulation results show that the algorithm effectively improves the packet delivery rate and reduces the delay of packets.

Structured datasets are difficult to keep up-to-date since the underlying facts evolve over time; curated data about business financials, organizational hierarchies, or drug interactions are constantly changing. Drafty is a platform that enlists visitors of an editable dataset to become ``user-editors'' to help solve this problem. It records and analyzes user-editors' within-page interactions to construct user interest profiles, creating a cyclical feedback mechanism that enables Drafty to target requests for specific corrections from user-editors. To validate the automatically generated user interest profiles, we surveyed participants who performed self-created tasks with Drafty and found their user interest score was 3.2 higher on data they were interested in versus data they had no interest in. Next, a 7-month live experiment compared the efficacy of user-editor corrections depending on whether they were asked to review data that matched their interests. Our findings suggest that user-editors are approximately 3 times more likely to provide accurate corrections for data matching their interest profiles, and about 2 times more likely to provide corrections in the first place.