This paper proposes a novel robust covert integrated sensing and communication (RC-ISAC) system, where mobile Warden tracking is leveraged to assist covert communication design. We focus on a typically overlooked yet highly threatening Warden-blocked scenario, in which temporary tracking loss prevents timely updates of covert communication strategy. To overcome this challenge, the reconfigurable intelligent surface (RIS) is introduced to establish a controllable sensing link that bypasses the obstacle. Furthermore, a robust extended Kalman filtering (R-EKF) strategy with a sensing-failure fallback mechanism is developed to achieve reliable Warden tracking, where sensing failures are detected and promptly addressed through re-scanning of the Warden. In addition, a high-capacity covert optimization (HCO) scheme is proposed to improve the covert transmission performance and maintain reliable Warden tracking, which is achieved by the joint design of ISAC sensing-communication beamforming and RIS passive beamforming. Simulation results demonstrate that the proposed RC-ISAC system achieves superior robustness and covert transmission performance compared with the no-RIS baseline and the element-wise optimization baseline.

Transformer has been very successful in various computer vision tasks and understanding the working mechanism of transformer is important. As touchstones, weakly-supervised semantic segmentation (WSSS) and class activation map (CAM) are useful tasks for analyzing vision transformers (ViT). Based on the plain ViT pre-trained with ImageNet classification, we find that multi-layer, multi-head self-attention maps can provide rich and diverse information for weakly-supervised semantic segmentation and CAM generation, e.g., different attention heads of ViT focus on different image areas and object categories. Thus we propose a novel method to end-to-end estimate the importance of attention heads, where the self-attention maps are adaptively fused for high-quality CAM results that tend to have more complete objects. Besides, we propose a ViT-based gradient clipping decoder for online retraining with the CAM results efficiently and effectively. Furthermore, the gradient clipping decoder can make good use of the knowledge in large-scale pre-trained ViT and has a scalable ability. The proposed plain Transformer-based Weakly-supervised learning method (WeakTr) obtains the superior WSSS performance on standard benchmarks, i.e., 78.5% mIoU on the $val$ set of PASCAL VOC 2012 and 51.1% mIoU on the $val$ set of COCO 2014. Source code and checkpoints are available at https://github.com/hustvl/WeakTr.

This paper investigates the reliability enhancement of integrated access and backhaul (IAB) networks in urban environments by jointly leveraging reconfigurable intelligent surfaces (RIS) and unmanned aerial vehicles (UAVs). We propose a RIS and UAV collaboration-enhanced IAB (RUC-IAB) network, where UAVs serve as mobile IAB nodes and the RIS is employed to establish robust line-of-sight (LoS) backhaul links. Our collaborative approach effectively mitigates both blockage-induced and signal-to-noise ratio (SNR)-limited outages, which are the two primary factors compromising transmission reliability in urban IAB networks. To further reduce the outages caused by data accumulation at the IAB node, we develop a joint UAV deployment and RIS beamforming optimization (URO) scheme to balance the access and backhaul transmission rates. In this scheme, a closed-form lower bound on the non-outage probability is derived to facilitate low-complexity UAV placement, and a semidefinite relaxation (SDR)-based method is proposed to optimize the RIS phase shifts. Simulation results show that the proposed URO scheme achieves a 44.72% reduction in average outage probability compared to the phase-alignment-based scheme across various backhaul distances.

Deep reinforcement Learning (DRL) offers a powerful framework for training AI agents to coordinate with human partners. However, DRL faces two critical challenges in human-AI coordination (HAIC): sparse rewards and unpredictable human behaviors. These challenges significantly limit DRL to identify effective coordination policies, due to its impaired capability of optimizing exploration and exploitation. To address these limitations, we propose an innovative behavior- and context-aware reward (BCR) for DRL, which optimizes exploration and exploitation by leveraging human behaviors and contextual information in HAIC. Our BCR consists of two components: (i) A novel dual intrinsic rewarding scheme to enhance exploration. This scheme composes an AI self-motivated intrinsic reward and a human-motivated intrinsic reward, which are designed to increase the capture of sparse rewards by a logarithmic-based strategy; and (ii) A new context-aware weighting mechanism for the designed rewards to improve exploitation. This mechanism helps the AI agent prioritize actions that better coordinate with the human partner by utilizing contextual information that can reflect the evolution of learning. Extensive simulations in the Overcooked environment demonstrate that our approach can increase the cumulative sparse rewards by approximately 20%, and improve the sample efficiency by around 38% compared to state-of-the-art baselines.

Camera and LiDAR serve as informative sensors for accurate and robust autonomous driving systems. However, these sensors often exhibit heterogeneous natures, resulting in distributional modality gaps that present significant challenges for fusion. To address this, a robust fusion technique is crucial, particularly for enhancing 3D object detection. In this paper, we introduce a dynamic adjustment technology aimed at aligning modal distributions and learning effective modality representations to enhance the fusion process. Specifically, we propose a triphase domain aligning module. This module adjusts the feature distributions from both the camera and LiDAR, bringing them closer to the ground truth domain and minimizing differences. Additionally, we explore improved representation acquisition methods for dynamic fusion, which includes modal interaction and specialty enhancement. Finally, an adaptive learning technique that merges the semantics and geometry information for dynamical instance optimization. Extensive experiments in the nuScenes dataset present competitive performance with state-of-the-art approaches. Our code will be released in the future.

Blockchain has been widely adopted for reputation management in Industrial Internet of Things (IIoT) systems due to its decentralization, immutability, and traceability. As more IIoT devices participate in reputation evaluations, there is an increasing demand for low-latency and reliable traceability of reputation data. However, conventional blockchain traceability queries require traversing all on-chain transactions to find target data, leading to high latency in both data querying and result verifying. To this end, we propose a fast and verifiable tracing approach for blockchain-enabled reputation management in IIoT networks, namely BCTree. Specifically, to reduce query time for reputation traceability, we accelerate multiple types of reputation queries by designing a hierarchical index structure for BCTree. This index structure incorporates Merkle B+-trees for fast retrieval of reputation values and Shifting Hash Bloom Filters (SHBFs) for fast retrieval of evaluators and evaluated entities. Furthermore, to reduce verification time for query result integrity, we develop a lightweight verification technique for BCTree based on a One-Hashing Bloom Filter (OHBF). Simulation results demonstrate that, compared to the state-of-the-art baseline, our BCTree reduces traceability query time by 86.78%, verification time by 22.01%, and verification object size by 33.22%, making it well-suited for latency-sensitive IIoT reputation management applications.

In this paper, we propose a hierarchical optimization approach that guarantees the maximum age of information (AoI) for uncrewed aerial vehicle (UAV) assisted Internet-of-Things (IoT) data collection. Our model is based on an energy-efficient simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) beamforming model. We formulate the optimization to minimize the UAV flight energy consumption subject to a maximum average AoI threshold by optimizing the UAV trajectory, IoT device scheduling, and STAR-RIS beamforming. To solve this, we develop an optimization-based hierarchical deep reinforcement learning (OH-DRL) algorithm that decomposes the formulated problem into an inter-cluster UAV visiting policy and STAR-RIS-based intra-cluster IoT scheduling policy. In OH-DRL, we jointly optimize the two policies in a high-level loop and a low-level loop, respectively. In the high-level loop, we design an AoI-guided DRL algorithm to determine the AoI-guaranteed UAV hovering position with minimal flight distance. In the low-level loop, a semidefinite relaxation (SDR)-based optimization algorithm further reduces the UAV’s flying time by minimizing the average AoI. Simulation results validate that OH-DRL achieves better convergence performance and energy-saving efficiency across different network scales. Compared to the state-of-the-art DRL algorithm, OH-DRL reduces the UAV flight energy consumption by 14.4% and decreases the number of training episodes required for convergence by 66%

Although Vision Transformer (ViT) has achieved significant success in computer vision, it does not perform well in dense prediction tasks due to the lack of inner-patch information interaction and the limited diversity of feature scale. Most existing studies are devoted to designing vision-specific transformers to solve the above problems, which introduce additional pre-training costs. Therefore, we present a plain, pre-training-free, and feature-enhanced ViT back-bone with Convolutional Multi-scale feature interaction, named ViT-CoMer, which facilitates bidirectional interaction between CNN and transformer. Compared to the state-of-the-art, ViT-CoMer has the following advantages: (1) We inject spatial pyramid multi-receptive field convolutional features into the ViT architecture, which effectively alleviates the problems of limited local information interaction and single-feature representation in ViT. (2) We propose a simple and efficient CNN- Transformer bidirectional fusion interaction module that performs multi-scale fusion across hierarchical features, which is beneficial for han-dling dense prediction tasks. (3) We evaluate the performance of ViT-CoMer across various dense prediction tasks, different frameworks, and multiple advanced pre-training. Notably, our ViT-CoMer-L achieves 64.3% AP on COCO val2017 without extra training data, and 62.1% mIoU on ADE20K val, both of which are comparable to state-of-the-art methods. We hope ViT-CoMer can serve as a new backbone for dense prediction tasks to facilitate future research. The code will be released at https://github.com/Traffic-xlviT-CoMer.

Network slicing (NS) is a promising technology that supports diverse requirements for next-generation low-latency wireless communication networks. However, the tampering attack is a rising issue of jeopardizing NS service-provisioning. To resist tampering attacks in NS networks, we propose a novel optimization framework for reliable NS resource allocation in a blockchain-secured low-latency wireless network, where trusted base stations (BSs) with high reputations are selected for blockchain management and NS service-provisioning. For such a blockchain-secured network, we consider that the latency is measured by the summation of blockchain management and NS service-provisioning, whilst the NS reliability is evaluated by the BS denial-of-service (DoS) probability. To satisfy the requirements of both the latency and reliability, we formulate a constrained computing resource allocation optimization problem to minimize the total processing latency subject to the BS DoS probability. To efficiently solve the optimization, we design a constrained deep reinforcement learning (DRL) algorithm, which satisfies both latency and DoS probability requirements by introducing an additional critic neural network. The proposed constrained DRL further solves the issue of high input dimension by incorporating feature engineering technology. Simulation results validate the effectiveness of our approach in achieving reliable and low-latency NS service-provisioning in the considered blockchain-secured wireless network.

Weakly-supervised visual recognition using inexact supervision is a critical yet challenging learning problem. It significantly reduces human labeling costs and traditionally relies on multi-instance learning and pseudo-labeling. This paper introduces WeakSAM and solves the weakly-supervised object detection (WSOD) and segmentation by utilizing the pre-learned world knowledge contained in a vision foundation model, i.e., the Segment Anything Model (SAM). WeakSAM addresses two critical limitations in traditional WSOD retraining, i.e., pseudo ground truth (PGT) incompleteness and noisy PGT instances, through adaptive PGT generation and Region of Interest (RoI) drop regularization. It also addresses the SAM's shortcomings of requiring human prompts and category unawareness in object detection and segmentation. Our results indicate that WeakSAM significantly surpasses previous state-of-the-art methods in WSOD and WSIS benchmarks with large margins, i.e. average improvements of 7.4% and 8.5%, respectively.

Whilst blockchain technology holds promise for secure Internet of Things (IoT) data management, its deployment in the massive Internet of Unmanned Aerial Vehicles (IoUAV) still faces significant challenges to satisfy strict requirements for low-latency query services and cost-effective resource consumption. To address these challenges, we present a lightweight multi-type data (MTD) blockchain architecture called LMChain with cost-effective MTD block scheduling. Specifically, LMChain incorporates cross-layer MTD blocks, wherein resource-constrained UAVs retain only lightweight block headers. Block bodies with high query probability are stored in fog nodes, while others are offloaded to cloud storage. Based on the MTD block structure, we develop a cost-effective block scheduling scheme to minimize the overall cost associated with LMChain storage and querying. A cooperative deep reinforcement learning (CDRL) algorithm is designed to efficiently schedule MTD blocks between the fog and cloud layers. Simulation results show that our LMChain significantly reduces the IoUAV blockchain system’s storage resource requirements and overall cost while supporting low-latency query services, making it well-suited for massive IoUAV applications.

Large language models (LLMs) demonstrate an impressive ability to internalize knowledge and answer natural language questions. Although previous studies validate that LLMs perform well on general knowledge while presenting poor performance on long-tail nuanced knowledge, the community is still doubtful about whether the traditional knowledge graphs should be replaced by LLMs. In this paper, we askif the schema of knowledge graph (i.e., taxonomy) is made obsolete by LLMs.Intuitively, LLMs should perform well on common taxonomies and at taxonomy levels that are common to people. Unfortunately, there lacks a comprehensive benchmark that evaluates the LLMs over a wide range of taxonomies from common to specialized domains and at levels from root to leaf so that we can draw a confident conclusion. To narrow the research gap, we constructed a novel taxonomy hierarchical structure discovery benchmark named TaxoGlimpse to evaluate the performance of LLMs over taxonomies. TaxoGlimpse covers ten representative taxonomies from common to specialized domains with in-depth experiments of different levels of entities in this taxonomy from root to leaf. Our comprehensive experiments of eighteen LLMs under three prompting settings validate that LLMs perform miserably poorly in handling specialized taxonomies and leaf-level entities. Specifically, the QA accuracy of the best LLM drops by up to 30% as we go from common to specialized domains and from root to leaf levels of taxonomies.

This paper proposes a blockchain-secured deep reinforcement learning (BC-DRL) optimization framework for data management and resource allocation in decentralized wireless mobile edge computing (MEC) networks. In our framework, we design a low-latency reputation-based proof-of-stake (RPoS) consensus protocol to select highly reliable blockchain-enabled BSs to securely store MEC user requests and prevent data tampering attacks. We formulate the MEC resource allocation optimization as a constrained Markov decision process that balances minimum processing latency and denial-of-service (DoS) probability. We use the MEC aggregated features as the DRL input to significantly reduce the high-dimensionality input of the remaining service processing time for individual MEC requests. Our designed constrained DRL effectively attains the optimal resource allocations that are adapted to the dynamic DoS requirements. We provide extensive simulation results and analysis to validate that our BC-DRL framework achieves higher security, reliability, and resource utilization efficiency than benchmark blockchain consensus protocols and MEC resource allocation algorithms.

In this paper, we develop a deep learning-based bandwidth allocation policy that is: 1) scalable with the number of users and 2) transferable to different communication scenarios, such as non-stationary wireless channels, different quality-of-service (QoS) requirements, and dynamically available resources. To support scalability, the bandwidth allocation policy is represented by a graph neural network (GNN), with which the number of training parameters does not change with the number of users. To enable the generalization of the GNN, we develop a hybrid-task meta-learning (HML) algorithm that trains the initial parameters of the GNN with different communication scenarios during meta-training. Next, during meta-testing, a few samples are used to fine-tune the GNN with unseen communication scenarios. Simulation results demonstrate that our HML approach can improve the initial performance by 8.79%, and sample efficiency by 73%, compared with existing benchmarks. After fine-tuning, our near-optimal GNN-based policy can achieve close to the same reward with much lower inference complexity compared to the optimal policy obtained using iterative optimization. Numerical results validate that our HML can reduce the computation time by approximately 200 to 2000 times than the optimal iterative algorithm.

Utilizing infrastructure and vehicle-side information to track and forecast the behaviors of surrounding traffic participants can significantly improve decision-making and safety in autonomous driving. However, the lack of real-world sequential datasets limits research in this area. To address this issue, we introduce V2X-Seq, the first large-scale sequential V2X dataset, which includes data frames, trajectories, vector maps, and traffic lights captured from natural scenery. V2X-Seq comprises two parts: the sequential perception dataset, which includes more than 15,000 frames captured from 95 scenarios, and the trajectory forecasting dataset, which contains about 80,000 infrastructure-view scenarios, 80,000 vehicle-view scenarios, and 50,000 cooperative-view scenarios captured from 28 intersections' areas, covering 672 hours of data. Based on V2X-Seq, we introduce three new tasks for vehicle-infrastructure cooperative (VIC) autonomous driving: VIC3D Tracking, Online-VIC Forecasting, and Offline-VIC Forecasting. We also provide benchmarks for the introduced tasks. Find data, code, and more up-to-date information at https://github.com/AIR-THU/DAIR-V2X-Seq.

In this article, we present practical stochastic modeling and detailed performance analysis of our double blockchain (DBC) from Haoet al.(2021) for secure information and reputation data management in large-scale wireless Internet of Things (IoT) networks. Specifically, the DBC is a private blockchain deployed on a cloud-fog communication network which is composed of an information blockchain (IBC) storing large amounts of IoT data in the cloud layer and a reputation blockchain (RBC) storing reputation data of the IoT devices in the near-terminal fog layer. The locations of the fog layer nodes are modeled according to a random Poisson point process (PPP) over a given 2-D area to approximate the stochastic property of real-world wireless node deployments. Furthermore, we assume that the number of IoT devices transmitting to the fog nodes also follow a random Poisson distribution. Based on these models, we derive novel closed-form expressions for the storage size, transmission latency, and tampering time of the IoT fog nodes in our DBC architecture. Numerical simulations highlight high storage scalability, low latency, and superior security of the DBC design, and provide insights into the performance gains for different fog node and IoT device densities.

Cross-modality person re-identification is a challenging task due to large cross-modality discrepancy and intramodality variations. Currently, most existing methods focus on learning modality-specific or modality-shareable features by using the identity supervision or modality label. Different from existing methods, this paper presents a novel Modality Confusion Learning Network (MCLNet). Its basic idea is to confuse two modalities, ensuring that the optimization is explicitly concentrated on the modality-irrelevant perspective. Specifically, MCLNet is designed to learn modality-invariant features by simultaneously minimizing inter-modality discrepancy while maximizing cross-modality similarity among instances in a single framework. Furthermore, an identity-aware marginal center aggregation strategy is introduced to extract the centralization features, while keeping diversity with a marginal constraint. Finally, we design a camera-aware learning scheme to enrich the discriminability. Extensive experiments on SYSU-MM01 and RegDB datasets show that MCLNet outperforms the state-of-the-art by a large margin. On the large-scale SYSU-MM01 dataset, our model can achieve 65.40 % and 61.98 % in terms of Rank-1 accuracy and mAP value.

Accurate mass segmentation on mammograms is a critical step in computer-aided diagnosis (CAD) systems. It is also a challenging task since some of the mass lesions are embedded in normal tissues and possess poor contrast or ambiguous margins. Besides, the shapes and densities of masses in mammograms are various. In this paper, a hybrid method combining a random walks algorithm and Chan-Vese (CV) active contour is proposed for automatic mass segmentation on mammograms. The data set used in this study consists of 1095 mass regions of interest (ROIs). First, the original ROI is preprocessed to suppress noise and surrounding tissues. Based on the preprocessed ROI, a set of seed points is generated for initial random walks segmentation. Afterward, an initial contour of mass and two probability matrices are produced by the initial random walks segmentation. These two probability matrices are used to modify the energy function of the CV model for prevention of contour leaking. Lastly, the final segmentation result is derived by the modified CV model, during which the probability matrices are updated by inserting several rounds of random walks. The proposed method is tested and compared with other four methods. The segmentation results are evaluated based on four evaluation metrics. Experimental results indicate that the proposed method produces more accurate mass segmentation results than the other four methods.

High level synthesis transformations play a major part in shaping the properties of the final circuit. However, most optimizations are performed without much knowledge of the final circuit layout. In this paper, we present a physically aware design flow for mapping high level application specifications to a synthesizable register transfer level hardware description. We study the problem of optimizing the data communication of the variables in the application specification. Our algorithm uses floorplan information that guides the optimization. We develop a simple, yet effective, incremental floorplanner to handle the perturbations caused by the data communication optimization. We show that the proposed techniques can reduce the wirelength of the final design, while maintaining a legal floorplan with the same area as the initial floorplan.

Rectangular cells can be flipped in place along either horizontal or vertical axis without changing the area of a layout. During floorplanning, both the location and orientation of cells are determined. However, the complexity of the floorplanning process usually means that the wirelength is not minimum. This paper proposes a technique for wirelength minimization based on in-place flipping of cells that can be applied to any floorplan style consisting of rectangular blocks or sub-blocks. Instead of conventional search procedures, a Boolean symbolic approach is proposed to generate flip-optimal floorplans. Experimental results show that it can effectively reduce the wirelength of current state of the art approaches, at no cost in area and with modest runtimes.