In this paper, we present a Spatio-Temporal Graph with Deep Embedded Clustering framework (STaG-DEC) that jointly detects, localizes, and classifies hybrid threats. It employs a Graph Convolutional Attention Network (GCAT) to extract spatial dependencies and fuses the resulting embeddings with temporally differenced measurements via concatenation and linear projection before the Transformer encoder, which suppresses slow operational drift, highlights abrupt attack-induced transients, and preserves long-range dependencies. To address class imbalance and overlapping patterns, standard cross-entropy with a DEC-based joint loss is proposed that promotes compact and better-separated latent clusters and stabilizes minority-class decision boundaries, thereby enabling precise node-level localization and classification across varying attack distributions. Experiments on IEEE 14-bus system demonstrate our superior detection performance with notably improved replay detection and robust behavior under imbalanced conditions.

With the rapid growth of cloud computing, securely outsourcing sensitive data has become crucial for protecting user privacy. SQL, renowned for its expressiveness and flexibility, remains the predominant query language for big data analytics. However, existing access control mechanisms for encrypted SQL databases primarily operate at coarse granularity levels, such as databases, tables, or columns, failing to meet nuanced user access requirements at the individual data-element level. To address this challenge, we propose EGuardSQL, a novel architecture that enables fine-grained, element-level access control over encrypted SQL tables. Our approach leverages Ciphertext-Policy Attribute-Based Encryption (CP-ABE) to define hybrid access policies that integrate both row-level and column-level permissions, facilitating precise and flexible control. To mitigate the inherent computational overhead associated with CP-ABE, particularly when dealing with complex policy structures, we introduce an optimized reusable sub-policy mechanism that effectively reduces redundant cryptographic computations during encryption and decryption processes. We formally prove the selective IND-CPA security of the EGuardSQL scheme under the decisional $q$-parallel Bilinear Diffie-Hellman Exponent (BDHE) assumption and demonstrate its resistance to collusion attacks. Comprehensive performance evaluations illustrate that EGuardSQL achieves significant improvements in computational and communication efficiency compared to conventional CP-ABE approaches, while maintaining low storage overhead and robust security guarantees.

In this paper, we propose SEA, a novel approach for active robot exploration through semantic map prediction and a reinforcement learning-based hierarchical exploration policy. Unlike existing learning-based methods that rely on one-step waypoint prediction, our approach enhances the agent's long-term environmental understanding to facilitate more efficient exploration. We propose an iterative prediction-exploration framework that explicitly predicts the missing areas of the map based on current observations. The difference between the actual accumulated map and the predicted global map is then used to guide exploration. Additionally, we design a novel reward mechanism that leverages reinforcement learning to update the long-term exploration strategies, enabling us to construct an accurate semantic map within limited steps. Experimental results demonstrate that our method significantly outperforms state-of-the-art exploration strategies, achieving superior coverage ares of the global map within the same time constraints.

Mirror detection aims to identify mirror areas in a scene, with recent methods either integrating depth information (RGB-D) or making use of temporal information (video). However, utilizing both data is still under-explored due to the lack of a high-quality dataset and an effective method for the RGB-D Video Mirror Detection (DVMD) problem. To the best of our knowledge, this is the first work to address the DVMD problem. To exploit depth and temporal information in mirror segmentation, we first construct a large-scale RGB-D Video Mirror Detection Dataset (DVMD-D), which contains 17977 RGB-D images from 273 diverse videos. We further develop a novel model, named DVMDNet, which can first locate the mirrors based on triple consistencies: local consistency, cross-modality consistency and global consistency, and then refine the mirror boundaries through content discontinuity, taking the temporal information within videos into account. We conduct a comparative study on the DVMD dataset, evaluating 12 state-of-the-art models (including single-image mirror detection, single-image glass detection, RGB-D mirror detection, video shadow detection, video glass detection, and video mirror detection methods). Code is available from https://github.com/UpChen/2025_DVMDNet.

Mapless navigation refers to the task of searching for a collision free path without relying on a pre-defined map. Most current works of mapless navigation assume accurate ground-truth localisation is available. However, this is not true, especially for indoor environments, where simultaneous localisation and mapping (SLAM) is needed for location estimation, which highly relies on the richness of environment features. In this work, we propose a novel deep reinforcement learning (DRL) based mapless navigation method without relying on the assumption of the availability of localisation. Our method utilises RGB-D based Oriented FAST and Rotated BRIEF (ORB) SLAM2 for robot localisation. Our policy effectively guides the robot's movement towards the target while enhancing robot pose estimation by considering the quality of the observed features along the selected paths. To facilitate policy training, we propose a compact state representation based on the spatial distributions of map points, which enhances the robot's awareness of areas with reliable map points. Furthermore, we suggest incorporating the relative pose error into the reward function. In this way, the policy will be more responsive to each single action. In addition, rather than utilising a pre-set threshold, we adopt a dynamic threshold to improve the policy's adaptability to variations in SLAM performance across different environments. The experiments in localisation challenging environments have demonstrated the remarkable performance of our proposed method. It outperforms the related DRL based methods in terms of success rate.

The high performance of tree ensemble classifiers benefits from a large set of rules, which, in turn, makes the models hard to understand. To improve interpretability, existing methods extract a subset of rules for approximation using model reduction techniques. However, by focusing on the reduced rule set, these methods often lose fidelity and ignore anomalous rules that, despite their infrequency, play crucial roles in real-world applications. This paper introduces a scalable visual analysis method to explain tree ensemble classifiers that contain tens of thousands of rules. The key idea is to address the issue of losing fidelity by adaptively organizing the rules as a hierarchy rather than reducing them. To ensure the inclusion of anomalous rules, we develop an anomaly-biased model reduction method to prioritize these rules at each hierarchical level. Synergized with this hierarchical organization of rules, we develop a matrix-based hierarchical visualization to support exploration at different levels of detail. Our quantitative experiments and case studies demonstrate how our method fosters a deeper understanding of both common and anomalous rules, thereby enhancing interpretability without sacrificing comprehensiveness.

This paper presents MirrorSAM2, the first framework that adapts Segment Anything Model 2 (SAM2) to the task of RGB-D video mirror segmentation. MirrorSAM2 addresses key challenges in mirror detection, such as reflection ambiguity and texture confusion, by introducing four tailored modules: a Depth Warping Module for RGB and depth alignment, a Depth-guided Multi-Scale Point Prompt Generator for automatic prompt generation, a Frequency Detail Attention Fusion Module to enhance structural boundaries, and a Mirror Mask Decoder with a learnable mirror token for refined segmentation. By fully leveraging the complementarity between RGB and depth, MirrorSAM2 extends SAM2's capabilities to the prompt-free setting. To our knowledge, this is the first work to enable SAM2 for automatic video mirror segmentation. Experiments on the VMD and DVMD benchmark demonstrate that MirrorSAM2 achieves SOTA performance, even under challenging conditions such as small mirrors, weak boundaries, and strong reflections.

Photorealistic style transfer aims to adapt the style of a reference image to a content image while preserving photorealism. Existing methods primarily rely on downsampling auto-encoders, which sacrifices spatial details due to the reduced feature map resolution and leads to artifacts. To address this problem, we propose a Resolution-preserving nEtwork for Style Transfer (REST) to maintain full spatial resolution during stylization. Our framework employs a resolution-preserving (RP) network (RPNet) that retains input resolution, effectively preserving fine details. Although RP features capture intricate spatial information, their reliance on high-resolution processing can limit semantic depth. To enhance feature representation, we introduce a semantic distillation module that transfers hierarchical patterns from VGG features into RPNet, ensuring balanced detail and semantic fidelity. Combined with advanced style transfer blocks, REST achieves photorealistic results with improved efficiency. Experiments show improved performance and faster processing over existing photorealistic style transfer approaches.

As the demand for asymmetric encryption in embedded devices increases, the SM2 algorithm, which is based on elliptic curve cryptography, has been widely applied in fields such as secure communication and digital signatures. Nevertheless, the intricate computations involved in elliptic curve point multiplication and modular arithmetic present substantial challenges to the resource-constrained and computationally limited nature of embedded systems. This paper presents several optimizations for the specific curve used in the SM2 algorithm, including a multilevel Karatsuba-Ofman large-number multiplier design, an enhanced fast modular reduction algorithm, and precomputed addition chains to accelerate modular inversion. With these optimizations, we achieve a high-performance, low-resource SM2 implementation, completing point multiplication in approximately 27.3k clock cycles. The optimized design was implemented on three Xilinx FPGA platforms: Virtex-7, Kintex-7, and Artix-7, with latencies of 0.238 ms, 0.254 ms, and 0.281 ms, respectively, and a DSP unit consumption of only 81. Compared with traditional implementations, the latency was reduced by approximately 15-20

Multi-class part parsing is a dense prediction task that segments objects into semantic components with multi-level abstractions. Despite its significance, this task remains challenging due to ambiguities at both part and class levels. In this paper, we propose a network that incorporates multi-class boundaries to precisely identify and emphasize the spatial boundaries of part classes, thereby improving segmentation quality. Additionally, we employ a weighted multi-label cross-entropy loss function to ensure balanced and effective learning from all parts. Experimental results validate the effectiveness of the proposed method, demonstrating its ability to enhance baseline performance on benchmark datasets.

False Data Injection (FDI) attack is one of critical threats to the reliability of power systems, where moving target defense (MTD) is an effective proactive detection defense strategy. Existing MTD studies have primarily concentrated on transmission networks through Distributed Flexible AC Transmission System (D-FACTS) devices. However, those methods cannot be extended to distribution networks due to significantly limited D-FACTS deployment. To detect FDI attacks in distribution networks, a topology-control-based moving target defense (TC-MTD) approach is proposed. Unlike D-FACTS-based MTD that alters line impedances, TC-MTD perturbs the system topology through switch operations and the rank of the admittance matrix difference is employed as a performance metric which indicates the extent of perturbation. Furthermore, a heuristic algorithm is developed to reconfigure the system topology, aiming to maximize detection performance and minimize power loss with the fewest switch operations. Case studies conducted on both modified and standard distribution networks validate the effectiveness of TC-MTD.

Most collection-based style transfer methods require training a separate model for each individual collection of styles, making the extension to multiple collections of styles less flexible. Besides, the existing collection-based methods are also less flexible in extending to new style collections in a continual manner. To address these issues, we propose a novel MultI-Dictionary Generative Adversarial Network framework (MID-GAN) for multi-collection style transfer. Specifically, we design a multi-dictionary architecture within a GAN, with each dictionary consisting of a set of local style codes for a specific style collection. Benefiting from the local style codes used in the dictionary, a stylization module with aligned skip connections is further proposed, which can better preserve both the local details and the overall image structure. The dictionary design allows a flexible extension to new style collections by readily adding new dictionaries and we propose a continual training strategy that can both preserve the style transfer ability of old styles and achieve good transfer results for newly added styles. Extensive experiments are performed to show that the proposed method is better than existing collection-based style transfer methods. We also demonstrate the proposed method can generate diverse meaningful style transfer results of the same style collection.

Diffusion generative models have gained rapid traction since 2020 due to their expressiveness and highquality outputs. Explaining and interpreting these models is essential for enabling further improvements and fostering trustworthiness. This systematic review identifies and analyzes interpretability methods applied to diffusion models across domains, highlighting key trends, outlining strategies, and identifying emerging research directions. We screened 1,489 papers published between 2020–2025 across IEEE, Scopus, DBLP, arXiv, and Elicit, and included 81 studies that met predefined criteria. Most methods target latent space analysis (n = 35), followed by data attribution (n = 16) and denoising dynamics (n = 14). Image generation and text-to-image synthesis dominate application areas (n = 73), with limited coverage in robotics, audio, and neuroscience (n = 8). This review offers a structured taxonomy, quantifies interpretability research trends, and identifies domain–specific and architectural gaps. Supplementary material and processing code are available at [GitHub link].

Picking up an entangled object is a difficult manipulation task due to its rich contact dynamics. Most existing solutions fail to produce grasp poses to enable reliable manipulation due to the dependence on simplified assumptions for the motion policies. Grasps generated by these methods tend to drop objects or cause undesired movements of non-grasped objects. To improve such object-disentangling tasks, we propose to extend the concept of reinforcement learning (RL)-based affordance to include arbitrary action consequences and implement a general affordance-based manipulation (GAM) framework. In the GAM, we train an RL agent that uses more fine-grained actions and outperforms previous methods with a smaller chance of dropping objects and making contact with non-grasped hooks. Then, a manipulation affordance prediction (MAP) model is trained to estimate the performances of the RL agent. Finally, the manipulation affordance-based grasp filter (MAGF) selects grasp poses that afford the desired manipulation performances, showing substantial improvements in five challenging hook disentangling tasks in simulation. The experiments show 1) the limitation of TAG generators, 2) the effectiveness of filtering TAGs with predicted manipulation performances based on the general affordance theory, and 3) the importance of avoiding contact with non-grasped objects in contact-rich manipulation.

This article is concerned with the secure strategy design for a class of discrete-time linear networked control systems subject to multiple cyber-attacks over a multi-channel framework. The multiple cyber-attacks discussed are groups of different attacks, which are random in sequence and can happen in any channel independently or simultaneously. No priority and duration limits are required for each type of attack in the multiple cyber-attacks, the dynamics of which are modeled by a Markovian chain. Moreover, a Bernoulli process is introduced to describe the channel conditions under attacks. A secure strategy based on Markovian-based observer is designed to guarantee the stability of the system with the ability to resist interference and adapt to multiple cyber-attacks. Finally, simulations are presented to verify the effectiveness of the proposed method.

Multi-class part parsing is a dense prediction task that decomposes objects into semantic components with multi-level abstractions. Despite the importance of this problem, it remains challenging due to the presence of both part-level and class-level ambiguities. In this paper, we propose GRPSNet network which integrates graph reasoning to capture relationships between parts for part segmentation. These captured relationships help to enhance the recognition and localization of parts. We also propose to exploit the relationships of part boundaries to further enhance the accuracy of part segmentation. The experimental results demonstrate the effectiveness of the proposed method and show that it achieves state-of-the-art performance on the benchmark datasets.

Label quality issues, such as noisy labels and imbalanced class distributions, have negative effects on model performance. Automatic reweighting methods identify problematic samples with label quality issues by recognizing their negative effects on validation samples and assigning lower weights to them. However, these methods fail to achieve satisfactory performance when the validation samples are of low quality. To tackle this, we develop Reweighter, a visual analysis tool for sample reweighting. The reweighting relationships between validation samples and training samples are modeled as a bipartite graph. Based on this graph, a validation sample improvement method is developed to improve the quality of validation samples. Since the automatic improvement may not always be perfect, a co-cluster-based bipartite graph visualization is developed to illustrate the reweighting relationships and support the interactive adjustments to validation samples and reweighting results. The adjustments are converted into the constraints of the validation sample improvement method to further improve validation samples. We demonstrate the effectiveness of Reweighter in improving reweighting results through quantitative evaluation and two case studies.

Hierarchical Reinforcement Learning (HRL) has shown superior performance for mapless navigation tasks. However, it remains limited in unstructured environments that might contain terrains like long corridors and dead corners, which can lead to local minima. This is because most HRL-based mapless navigation methods employ a simplified reward setting and exploration strategy. In this work, we propose a novel reward function for training the high-level (HL) policy, which contains two components: extrinsic reward and intrinsic reward. The extrinsic reward encourages the robot to move towards the target location, while the intrinsic reward is computed based on novelty, episode memory and memory decaying, making the agent capable of accomplishing spontaneous exploration. We also design a novel neural network structure that incorporates an LSTM network to augment the agent with memory and reasoning capabilities. We test our method in unknown environments and specific scenarios prone to the local minimum problem to evaluate the navigation performance and local minimum resolution ability. The results show that our method significantly increases the success rate when compared to advanced RL-based methods, achieving a maximum improvement of nearly 28%. Our method demonstrates effective improvement in addressing the local minimum issue, especially in cases where the baselines fail completely. Additionally, numerous ablation studies consistently confirm the effectiveness of our proposed reward function and neural network structure.

The rapid advancement in visual generation, particularly the emergence of pre-trained text-to-image and text-to-video models, has catalyzed growing interest in training-free video editing research. Mirroring training-free image editing techniques, current approaches preserve original video information through video input inversion and manipulating intermediate features and attention during the inference process to achieve content editing. Although they have demonstrated promising results, the lossy nature of the inversion process poses significant challenges in maintaining unedited regions of the video. Furthermore, feature and attention manipulation during inference can lead to unintended over-editing and face challenges in both local temporal continuity and global content consistency. To address these challenges, this study proposes a score distillation paradigm based on pre-trained text-to-video models, where the original video is iteratively optimized through multiple steps guided by editing gradients provided by score distillation to ultimately obtain the target video. The iterative optimization starting from the original video, combined with content preservation loss, ensures the maintenance of unedited regions in the original video and suppresses over-editing. To further guarantee video content consistency and temporal continuity, we additionally introduce a global consistency auxiliary loss and optical flow prediction-based local editing gradient smoothing. Experiments demonstrate that these strategies effectively address the aforementioned challenges, achieving comparable or superior performance across multiple dimensions including preservation of unedited regions, local temporal continuity, and global content consistency of editing results, compared to state-of-the-art methods.

Solving reinforcement learning (RL)-based mapless navigation tasks is challenging due to their sparse reward and long decision horizon nature. Hierarchical reinforcement learning (HRL) has the ability to leverage knowledge at different abstract levels and is thus preferred in complex mapless navigation tasks. However, it is computationally expensive and inefficient to learn navigation end-to-end from raw high-dimensional sensor data, such as Lidar or RGB cameras. The use of subgoals based on a compact intermediate representation is therefore preferred for dimension reduction. This work proposes an efficient HRL-based framework to achieve this with a novel scoring method, named Predictive Neighbouring Space Scoring (PNSS). The PNSS model estimates the explorable space for a given position of interest based on the current robot observation. The PNSS values for a few candidate positions around the robot provide a compact and informative state representation for subgoal selection. We study the effects of different candidate position layouts and demonstrate that our layout design facilitates higher performances in longer-range tasks. Moreover, a penalty term is introduced in the reward function for the high-level (HL) policy, so that the subgoal selection process takes the performance of the low-level (LL) policy into consideration. Comprehensive evaluations demonstrate that using the proposed PNSS module consistently improves performances over the use of Lidar only or Lidar and encoded RGB features Note to Practitioners —This paper seeks to improve robot mapless navigation capabilities where the robot is expected to navigate to a goal location without knowing the map of the environment. This ability is highly demanded in many applications that require autonomous operations in unstructured environments, including both indoor and outdoor scenarios, involving tasks such as service robots for domestic and public environments, logistics in industrial warehouses, urban search and rescue missions, and disaster relief efforts, where detailed and accurate maps are difficult to obtain in advance. In this work, we focus on reinforcement learning-based mapless navigation. It is known that such methods struggle in complex long-range tasks, e.g. stuck in a local region by multiple objects. Therefore, this paper proposes a novel mapless navigation method inspired by human navigation behaviours. We enable a robot to split a long-range navigation task into multiple segments, by selecting and navigating to short-term goals. These subgoals are selected each time from a number of candidate positions located around the robot. The process stops when the robot reaches the final target location. When selecting a short-term goal, we use a deep neural network to predict the openness around each candidate subgoal position, named the Predictive Neighbouring Space Scoring (PNSS), from raw images and Lidar scans. In addition, we study the effects of different arrangements of candidate subgoal locations and select the optimal one. Experiments conducted in photo-realistic simulation environments demonstrate the effectiveness of our method, showcasing superior performance over baselines. It is worth noting that our agent is only trained in domestic environments using the iGibson simulator. For applications in other environments, additional training in more representative settings specific to corresponding scenarios will be necessary. In the future, our intention is to validate our methods in complex real-world environments and narrow the simulation-to-reality gap for long-horizon navigation tasks.