Search Results (129)

Static honeypot deployment and one-shot attack-path analysis often become ineffective against adaptive adversaries because fixed decoy layouts are easy to fingerprint and risk estimates quickly go stale. This paper presents a unified, mathematically grounded TrapManager framework that couples graph representation learning with budget-constrained combinatorial optimization for dynamic cyber deception. We model attacker progression on vulnerability-based attack graphs and learn context-aware node embeddings using a Graph Attention Network (GAT) that fuses vulnerability-driven risk signals (e.g., CVSS-derived node scores) with structural features. The learned representations are used to estimate edge plausibility and rank candidate source–target routes at the path level. Given limited resources, we formulate pointTrap placement as a Mixed-Integer Programming (MIP) problem that maximizes the expected interception of high-risk paths while penalizing deployment cost under explicit budget constraints, including mandatory coverage of the top-ranked critical paths. To enable online adaptiveness, a pointTrap-triggered, event-driven feedback mechanism locally amplifies risk around alerted regions, updates path weights without retraining the GAT, and re-solves the MIP for rapid redeployment. Experiments on MulVAL-generated benchmark attack graphs and cross-domain transfer settings demonstrate fast convergence, strong discrimination between attack and non-attack edges, and early interception within a small number of hops even with minimal decoy budgets. Overall, the proposed framework provides a scalable and resource-efficient approach to closed-loop attack-path defense by integrating attention-based learning and integer optimization. Full article

General-purpose large language models excel at open-domain question answering, but in railway operation and maintenance (O&M) scenarios they still suffer from hallucinated knowledge and poor domain adaptation. In practice, railway O&M knowledge mainly arises from two heterogeneous sources: spatio-temporal data such as train trajectories, which are organized along the spatial layout of railway lines, and domain documents such as operating rules, which exhibit varying degrees of structural regularity. Traditional retrieval-augmented generation (RAG) systems usually flatten these multi-source data into a single unstructured text space and perform global retrieval in one embedding space, which easily introduces noisy context and makes it difficult to precisely target knowledge for specific lines, sections, or equipment states. To overcome these limitations, we propose CARAG, a context-aware RAG framework tailored to railway O&M data. CARAG treats domain documents and spatial data as a unified knowledge substrate and builds a spatial knowledge graph with concept and instance levels. On top of this knowledge graph, a GraphReAct-based multi-turn interaction mechanism guides the LLM to reason and act over the concept knowledge graph, dynamically navigating to spatially and semantically relevant candidate regions, within which vector retrieval and instance-level graph retrieval are performed. Experiments show that CARAG significantly outperforms baseline RAG methods on RAGAS metrics, confirming the effectiveness of structure-guided multi-step reasoning for question answering over multi-source heterogeneous railway O&M data. Full article

Title block compliance checking requires interpreting irregular tabular layouts and reporting structural inconsistencies, not only extracting metadata. This paper introduces a user-in-the-loop, template-based method that leverages a graphical annotation workflow to encode title block structure as a hierarchical annotation graph combining detected primitives (cells/text) with user-defined semantic entities (key–value pairs, tables, headers). The resulting template is matched onto target title blocks using relative positional constraints and category-specific rules that distinguish acceptable variability from non-compliance (e.g., variable-size tables versus missing fields). The system outputs extracted key–value information and localized warning logs for end-user correction. On a real industrial example from the nuclear domain, the approach achieves 98–99% compliant annotation matching and 84% accuracy in flagging structural/content deviations, while remaining tolerant to moderate layout changes. Limitations and extensions are discussed, including support for additional fields, improved key similarity metrics, operational deployment with integrated feedback and broader benchmarking. Full article

Cyberbullying and hate speech increasingly appear in multi-modal social media posts, where images and text are combined in diverse and fast changing ways across platforms. These posts differ in style, vocabulary and layout, and labeled data are sparse and noisy, which makes it difficult to train detectors that are both reliable and deployable under tight computational budgets. Many high performing systems rely on large vision language backbones, full parameter fine tuning, online retrieval or model ensembles, which raises training and inference costs. We present a parameter efficient cross-platform multi-modal transfer learning framework for cyberbullying and hateful content detection. Our framework has three components. First, we perform domain adaptive pretraining of a compact ViLT backbone on in domain image-text corpora. Second, we apply parameter efficient fine tuning that updates only bias terms, a small subset of LayerNorm parameters and the classification head, leaving the inference computation graph unchanged. Third, we use noise aware knowledge distillation from a stronger teacher built from pretrained text and CLIP based image-text encoders, where only high confidence, temperature scaled predictions are used as soft labels during training, and teacher models and any retrieval components are used only offline. We evaluate primarily on Hateful Memes and use IMDB as an auxiliary text only benchmark to show that the deployment aware PEFT + offline-KD recipe can still be applied when other modalities are unavailable. On Hateful Memes, our student updates only 0.11% of parameters and retain about 96% of the AUROC of full fine-tuning. Full article

Digital microfluidic biochips (DMFBs) find extensive applications in biochemical experiments, medical diagnostics, and safety-critical domains, with their reliability dependent on efficient online testing technologies. However, traditional random search algorithms suffer from slow convergence and susceptibility to local optima under complex fluidic constraints. This paper proposes a hybrid optimization method based on priority strategy and an improved sparrow search algorithm for DMFB online test path planning. At the algorithmic level, the improved sparrow search algorithm incorporates three main components: tent chaotic mapping for population initialization, cosine adaptive weights together with Elite Opposition-based Learning (EOBL) to balance global exploration and local exploitation, and a Gaussian perturbation mechanism for fine-grained refinement of promising solutions. Concurrently, this paper proposes an intelligent rescue strategy that integrates global graph-theoretic pathfinding, local greedy heuristics, and space–time constraint verification to establish a closed-loop decision-making system. The experimental results show that the proposed algorithm is efficient. On the standard 7 × 7–15 × 15 DMFB benchmark chips, the shortest offline test path length obtained by the algorithm is equal to the length of the Euler path, indicating that, for these regular layouts, the shortest test path has reached the known optimal value. In both offline and online testing, the shortest paths found by the proposed method are better than or equal to those of existing mainstream algorithms. In particular, for the 15 × 15 chip under online testing, the proposed method reduces the path length from 543 and 471 to 446 compared with the IPSO and IACA algorithms, respectively, and reduces the standard deviation by 53.14% and 39.4% compared with IGWO in offline and online testing. Full article

Indoor positioning using commodity WiFi has gained significant attention; however, achieving sub-meter accuracy across diverse layouts remains challenging due to multipath fading and Non-Line-Of-Sight (NLOS) effects. In this work, we propose a hybrid Graph–Temporal Convolutional Network (GTCN) model that incorporates Access Point (AP) geometry through graph convolutions while capturing temporal signal dynamics via dilated temporal convolutional networks. The proposed model adaptively learns per-AP importance using a lightweight gating mechanism and jointly exploits WiFi Received Signal Strength (RSS) and Round-Trip Time (RTT) features for enhanced robustness. The model is evaluated across four experimental areas such as lecture theatre, office, corridor, and building floor covering areas from 15 m × 14.5 m to 92 m × 15 m. We further analyze the sensitivity of the model to AP density under both LOS and NLOS conditions, demonstrating that positioning accuracy systematically improves with denser AP deployment, especially in large-scale mixed environments. Despite its high accuracy, the proposed GTCN remains computationally lightweight, requiring fewer than ${10}^5$ trainable parameters and only tens of MFLOPs per inference, enabling real-time operation on embedded and edge devices. Full article

The deparser stage in the Protocol-Independent Switch Architecture (PISA) is often overshadowed by parser and match-action optimizations. Yet, it remains a critical performance bottleneck in P4-programmable FPGA data planes. Challenges associated with the deparser stem from dynamic header layouts, variable emission orders, and alignment constraints, which often necessitate resource-intensive designs, such as wide, dynamic crossbar routing. While compile-time specialization techniques can reduce logic usage, they sacrifice runtime adaptability: any change to the protocol graph, including adding, removing, or reordering headers, requires full hardware resynthesis and re-implementation, limiting their practicality for evolving or multi-tenant workloads. This work presents a unified FPGA-targeted deparser architecture that merges templated and overlay concepts within a hardware–software co-design framework. At design time, template parameters define upper bounds on protocol complexity, enabling resource-efficient synthesis tailored to specific workloads. Within these bounds, runtime reconfiguration is supported through overlay control tables derived from static deparser DAG analysis, which capture the per-path emission order, header alignments, and offsets. These tables drive protocol-agnostic, chunk-based emission blocks that eliminate the overhead of crossbar interconnects, thereby significantly reducing complexity and resource usage. The proposed design sustains high throughput while preserving the flexibility needed for in-field updates and long-term protocol evolution. Full article

Industrial tourism has become an increasingly important means of promoting corporate identity and fostering public engagement, yet many factory tours suffer from fragmented layouts, congestion, and low visitor engagement. This study addresses these challenges by developing a hybrid framework that integrates expert-driven spatial zoning with bottom-up visitor analytics. Using an optical lens manufacturer in Taiwan as a case study, we applied a three-step process: (1) Delphi-based zoning of key subareas into functional zones, (2) empirical analysis of visitor movement, feedback, and shadowing data, and (3) computational spatial evaluation through Visibility Graph Analysis (VGA). The findings revealed thematic inconsistencies, overlooked exhibits, and bottlenecks that disrupted narrative flow and reduced engagement. Spatial reorganization—such as relocating interactive subareas to visually integrated zones—enhanced circulation, storytelling alignment, and experiential coherence. A complementary service blueprint linked spatial redesign to operational delivery, ensuring consistency between frontstage activities and backstage support. The data-driven spatial analytics validated the effectiveness of this study’s hybrid approach—combining expert-driven insights with grounded visitor behavior data—to optimize factory tours. Spatial efficiency contributes to reduced energy use and congestion, participatory experiences enhance education and inclusivity, and improved visitor satisfaction strengthens brand resilience and economic viability. The framework thus provides a replicable and sustainable model for industrial tourism development across diverse manufacturing sectors. Full article

Successful health promotion involves messages that are quickly captured and held long enough to permit eligibility, credibility, and calls to action to be coded. This research develops an exploratory eye-tracking atlas of breast cancer screening ads viewed by midlife women and a replicable pipeline that distinguishes early capture from long-term processing. Areas of Interest are divided into design-influential categories and graphed with two complementary measures: first hit and time to first fixation for entry and a tie-aware pairwise dominance model for dwell that produces rankings and an “early-vs.-sticky” quadrant visualization. Across creatives, pictorial and symbolic features were more likely to capture the first glance when they were perceptually dominant, while layouts containing centralized headlines or institutional cues deflected entry to the message and source. Prolonged attention was consistently focused on blocks of text, locations, and badges of authoring over ornamental pictures, demarcating the functional difference between capture and processing. Subgroup differences indicated audience-sensitive shifts: Older and household families shifted earlier toward source cues, more educated audiences shifted toward copy and locations, and younger or single viewers shifted toward symbols and images. Internal diagnostics verified that pairwise matrices were consistent with standard dwell summaries, verifying the comparative approach. The atlas converts the patterns into design-ready heuristics: defend sticky and early pieces, encourage sticky but late pieces by pushing them toward probable entry channels, de-clutter early but not sticky pieces to convert to processing, and re-think pieces that are neither. In practice, the diagnostics can be incorporated into procurement, pretesting, and briefs by agencies, educators, and campaign managers in order to enhance actionability without sacrificing segmentation of audiences. As an exploratory investigation, this study invites replication with larger and more diverse samples, generalizations to dynamic media, and associations with downstream measures such as recall and uptake of services. Full article

This research offers a fresh lens on Iranian cultural heritage houses by interrogating the overlooked role of Orosi windows in shaping socio-spatial accessibility and visual permeability. While these decorative stained-glass features are traditionally appreciated for their artistry and environmental performance, their functional impact on visibility and spatial interaction remains underexplored. The study aims to assess how window visual permeability influences socio-spatial accessibility within the hierarchical layouts of historic houses in Iran. To this end, a quantitative approach was adopted, applying convex space analysis to examine socio-spatial dynamics and visibility graph analysis (VGA) to study visual permeability within the space syntax framework. Fifteen heritage houses were analysed under two conditions using VGA: their current status quo, and a hypothetical model in which windows were treated as fully transparent, allowing unobstructed sightlines. The analyses demonstrated that removing window barriers enhanced visual integration and connectivity across all cases. Statistical t-tests further confirmed that these differences were significant, establishing that Orosi windows exert a profound influence on visual permeability. Beyond their ornamental and climatic roles, this study redefines Orosi windows as dynamic cultural devices that actively script human visibility, privacy, and interaction, revealing how historical design intelligence can inform sustainable, culturally responsive architectural practices. Full article

The perception of 3D space by mobile robots is rapidly moving from flat metric grid representations to hybrid metric-semantic graphs built from human-interpretable concepts. While most approaches first build metric maps and then add semantic layers, we explore an alternative, concept-first architecture in which spatial understanding emerges from asynchronous concept agents that directly instantiate and manage semantic entities. Our robot employs two spatial concepts—room and door—implemented as autonomous processes within a cognitive distributed architecture. These concept agents cooperatively build a shared scene graph representation of indoor layouts through active exploration and incremental validation. The key architectural principle is hierarchical constraint propagation: Room instantiation provides geometric and semantic priors to guide and support door detection within wall boundaries. The resulting structure is maintained by a complementary functional principle based on prediction-matching loops. This approach is designed to yield an actionable, human-interpretable spatial representation without relying on any pre-existing global metric map, supporting scalable operation and persistent, task-relevant understanding in structured indoor environments. Full article

Analyzing the spatial layout equilibrium of highway transportation networks is essential for optimizing transportation networks, enhancing system efficiency and sustainability. To promote the equitable distribution and management of highway traffic resources, this study introduces a framework for assessing the spatial layout equilibrium of highway networks based on community structure. A new algorithm, named the C-Louvain algorithm, is introduced in this paper to address improving the stability of detection results in unconnected networks. The method first constructs a spatial node-based network, then detects the community structure of the highway network using the C-Louvain algorithm, and identifies key communities of the community structure network through a depth-first search. Network spatial layout imbalance is quantitatively assessed through supply–demand equilibrium analysis based on the Gini coefficient. This methodology is applied to the regional highway network in Shenyang, China. Results indicate that the C-Louvain method is optimal, excelling in accuracy, volatility, and efficiency compared to the classic FN, Leiden, and Louvain algorithms, providing a valuable contribution to the literature on graph clustering and data mining. There are significant differences in the number of communities within different connected components, which reflects the heterogeneity of the network’s structure. By this method, the imbalanced area in the highway transportation network layout is quickly found, and the equitable distribution of traffic resources is quantitatively evaluated. The research results can provide a theoretical basis for managers to make scientific investment decisions for road network construction. Full article

We address issues in existing house layout generation methods, including chaotic room layouts, limited iterative refinement, and restricted style diversity. We propose Graph-RWGAN, a generative adversarial network based on a multi-relational graph attention mechanism, to automatically generate reasonable and globally consistent house layouts under weak constraints. In our framework, rooms are represented as graph nodes with semantic attributes. Their spatial relationships are modeled as edges. Optional room-level objects can be added by augmenting node attributes. This allows for object-aware layout generation when needed. The multi-relational graph attention mechanism captures complex inter-room relationships. Iterative generation enables stepwise layout optimization. Fusion of node features with building boundaries ensures spatial accuracy and structural coherence. A conditional graph discriminator with Wasserstein loss constrains global consistency. Experiments on the RPLAN dataset show strong performance. FID is 92.73, SSIM is 0.828, and layout accuracy is 85.96%. Room topology accuracy reaches 95%, layout quality 90%, and structural coherence 95%, outperforming House-GAN, LayoutGAN, and MR-GAT. Ablation studies confirm the effectiveness of each key component. Graph-RWGAN shows strong adaptability, flexible generation under weak constraints, and multi-style layouts. It provides an efficient and controllable scheme for intelligent building design and automated planning. Full article

Road-surface distress poses a serious threat to traffic safety and imposes a growing burden on urban maintenance budgets. While modern detectors based on convolutional networks and Vision Transformers achieve strong frame-level performance, they often overlook an essential property of road environments—structural symmetry within road networks and damage patterns. We present Graph-MambaRoadDet (GMRD), a symmetry-aware and lightweight framework that integrates dynamic graph reasoning with state–space modeling for accurate, topology-informed, and real-time road damage detection. Specifically, GMRD employs an EfficientViM-T1 backbone and two DefMamba blocks, whose deformable scanning paths capture sub-pixel crack patterns while preserving geometric symmetry. A superpixel-based graph is constructed by projecting image regions onto OpenStreetMap road segments, encoding both spatial structure and symmetric topological layout. We introduce a Graph-Generating State–Space Model (GG-SSM) that synthesizes sparse sample-specific adjacency in $\mathcal{O}\left(M\right)$ time, further refined by a fusion module that combines detector self-attention with prior symmetry constraints. A consistency loss promotes smooth predictions across symmetric or adjacent segments. The full INT8 model contains only 1.8 M parameters and 1.5 GFLOPs, sustaining 45 FPS at 7 W on a Jetson Orin Nano—eight times lighter and 1.7× faster than YOLOv8-s. On RDD2022, TD-RD, and RoadBench-100K, GMRD surpasses strong baselines by up to +6.1 mAP_50:95 and, on the new RoadGraph-RDD benchmark, achieves +5.3 G-mAP and +0.05 consistency gain. Qualitative results demonstrate robustness under shadows, reflections, back-lighting, and occlusion. By explicitly modeling spatial and topological symmetry, GMRD offers a principled solution for city-scale road infrastructure monitoring under real-time and edge-computing constraints. Full article

This study investigated how icon array layouts influence comprehension of medical risk information, particularly in relation to users’ cognitive abilities. In a within-subjects experiment (N = 121), participants reviewed clinical scenarios with treatment-related risks and side effect risks displayed in either separated or integrated icon arrays. Comprehension was significantly higher for separated treatment-related risk layouts (p < 0.001), while side effect layout showed no effect. Numeracy and graph literacy significantly predicted comprehension. Crucially, individuals with lower numeracy showed marked gains when viewing separated formats, whereas those with higher numeracy performed well regardless of layout. Despite this, participants preferred hybrid formats—separated treatment-related risk with integrated side effect risks—revealing a critical preference–performance gap. By demonstrating how visual layout interacts with user abilities, this study provides actionable guidance for patient decision aid design. The findings show that comprehension accuracy must take precedence over layout compactness and user preference, with separated layouts recommended for treatment-related risks—especially for individuals with lower numeracy—and greater flexibility allowed for side effect risks when space is limited. Full article