Search Results (51)

Infrared and visible image fusion needs to preserve visible texture details and infrared thermal saliency, yet emphasizing one modality may suppress or distort useful information from the other, while cross-modal differences may also contain noise, pseudo-textures, or locally incompatible boundaries. We propose ARC³Fusion, which reformulates image fusion as a reliability-calibrated consensus–complementarity–conflict process to achieve a more effective balance between visible texture detail and infrared target saliency. A progressive shared encoder and a modality-specific residual adapter first produce comparable yet modality-aware features. Cross-Modal Explainable Residual Decomposition then estimates jointly supported consensus and represents the information unexplained by the opposite modality as candidate residuals. Trustworthy Complementarity Verification evaluates infrared residuals using source intensity and edge evidence, while visible residuals are examined using source cues and learnable frequency-pattern evidence. Cross-Modal Conflict Estimation further characterizes local incompatibility through co-activation, reliability, amplitude imbalance, edge-strength mismatch, and orientation mismatch. Conflict-Aware Routing finally coordinates consensus and verified residuals according to these relation cues. Unlike conventional shared–private decomposition that directly preserves private features, ARC³Fusion treats modality-specific residuals as candidates that must be verified and conflict-coordinated before fusion. Experiments on LLVIP, MSRS, and TNO demonstrate consistent fusion performance. On LLVIP, ARC³Fusion achieves the best EN, SF, AG, VIF, and SCD values of 7.158, 14.467, 4.331, 1.136, and 1.229, respectively. These results indicate that verifying modality-specific residuals and coordinating local conflicts improves the joint preservation of visible texture details and infrared thermal saliency. Full article

Single-entity electrochemistry (SEE) expands the scope of analytical electrochemical measurement by shifting attention from ensemble-averaged currents to individually resolved stochastic events. This review evaluates progress toward two analytical endpoints, trustworthy detection of single molecules and context-preserving interrogation of single cells, with emphasis on quantitative rigor rather than platform novelty alone. Across nanoparticle collisions, nanopores, confined nanoelectrodes, vesicle electrochemical cytometry, intracellular nanopipettes, and array-enabled single-cell devices, the decisive analytical issue is no longer simply whether one entity can be detected, but whether event assignment, calibration, throughput, and reproducibility are sufficient to support credible inference. Representative primary studies are compared through shared metrics including event frequency, temporal resolution, bandwidth, molecular counts, detection limit, affinity, and effective yield of analyzable events. Particular attention is given to three recurring bottlenecks: interfacial variability, model-dependent event interpretation, and incomplete reporting of denominators such as rejected events, insertion success, and pore-to-pore or cell-to-cell reproducibility. The current evidence base is strongest in secretion and vesicle studies, whereas confinement-enabled and multimodal routes define the leading edge of single-molecule analysis. Overall, SEE is developing not as a single universal platform, but as a family of interface-controlled, data-rich analytical strategies whose future analytical value will depend on standardized reporting, multimodal validation, and benchmarking practices that preserve both sensitivity and confidence of assignment. Full article

There is an enormous rise in the amount of user-generated content on social media. That makes it easier for hateful and fake messages to spread, and threatens both societal stability and public trust in institutions. Most of current solutions have fundamental limitations due to modal limitations (i.e., each solution only uses one type of data at a time), lack of user context integration, poor synchronization across different types of data, and poor resilience to manipulation by adversaries. As a result, most solutions are subject to compound loss in terms of their ability to generalize well, classify correctly, or remain reliable when deployed in real-world environments. To address all of the above challenges, we propose a comprehensive and modular analytical framework consisting of five interconnected components that integrate contextual representation learning, multimodal semantic alignment, graph-based propagation modeling, adaptive inference, and consistency validation for hate speech and fake post detection. First is our Context-Driven Social Vector Extraction methodology, which provides enriched contextual embeddings by extracting and combining text-based metadata, image-based metadata, temporal metadata, and behavioral metadata. We use those embeddings in our second module, Multimodal Label Fusion via Mutual Co-Attention (CMF-MCA). Our CMF-MCA module incorporates two transformers with co-attention mechanisms that can mutually annotate text and images. In our third methodology, Semantic Propagation Graph for Hate and Fake Correlation (SPG-HFC), we implement a relational graph attention mechanism that captures both the influence of semantics and how communities propagate information about hate and fake posts. The fourth module, Adaptive Modality Routing via Reinforcement (AMR-R), routes based on the modality of the input and whether the input is simple enough to be classified using machine learning or complex enough to require deep learning. Finally, our Counterfactual Consistency Validation Engine (CCVE) is used after prediction to validate that the model’s predictions are consistent with the output data by creating counterfactuals and validating them. Therefore, in addition to improving the overall accuracy of hate speech and fake post detections, our proposed framework also improves its scalability and inference reliability. Additionally, because our framework allows multimodal classifications that include both context and behavior, it enables the scalable and trustworthy development of content moderation systems. Full article

Oil and gas pipelines are critical infrastructures that require continuous and reliable monitoring to detect leaks, pressure anomalies, corrosion, and unauthorized activities. Wireless sensor networks (WSNs) have emerged as an effective solution for large-scale pipeline monitoring due to their low deployment cost and real-time sensing capabilities. However, the resource-constrained nature of sensor nodes and the open wireless communication environment expose pipeline monitoring systems to various routing attacks, for example, blackhole, sinkhole, selective forwarding, and false data injection attacks, while simultaneously demanding strict energy efficiency to prolong network lifetime. In this paper, we propose SEER-PM (Secure and Energy-Efficient Routing for Pipeline Monitoring): a novel protocol that integrates an Artificial neural network (ANN)-based trust mechanism with energy-aware routing metrics. SEER-PM dynamically evaluates node trustworthiness based on packet forwarding behavior, residual energy, and signal consistency. By training the ANN on historical behavioral data, the system accurately detects malicious nodes with high precision. Simulation results demonstrate that SEER-PM outperforms existing secure routing protocols (Sec-AODV and T-LEACH) in terms of packet delivery ratio (PDR) by 14%, detection rate by 9.5%, and network lifetime by 12% under heavy attack scenarios. The proposed protocol enhances the reliability, security, and sustainability of pipeline monitoring WSNs operating in harsh and remote environments. Full article

Artificial intelligence systems increasingly mediate high-stakes human activities, yet urban navigation remains highly challenging for blind and visually impaired individuals. Although digital navigation technologies have significantly improved route planning and accessibility, many existing systems still rely on generic interaction paradigms that insufficiently account for cognitive load, contextual uncertainty, and the adaptive needs of vulnerable users. This challenge highlights the importance of Human-Centred AI approaches capable of supporting not only functional accessibility, but also cognitively sustainable and trustworthy interaction. This paper introduces LAZAR, a human-centred adaptive AI framework for accessible urban mobility grounded in a user-centred design methodology and formalised through a structured Software Requirements Specification. Rather than focusing exclusively on route optimisation, LAZAR approaches assistive navigation as an adaptive human–AI interaction problem in which instructional granularity, interaction frequency, and feedback mechanisms are designed to support user autonomy and situational awareness whilst limiting unnecessary cognitive burden. The proposed framework integrates high-fidelity prototyping, accessibility-oriented interaction modelling, and a modular multi-agent architecture intended to support adaptive and personalised guidance. Central to the approach is a cognitive load-aware interaction layer designed to regulate the presentation and timing of navigational assistance according to user needs and contextual conditions. The proposed multi-agent architecture is presented as a modular design framework whose interaction principles and interface logic were partially operationalised in the evaluated prototype. The complete integration of all adaptive coordination mechanisms, together with large-scale real-world validation, remains part of ongoing and future development work. This work contributes a structured methodology for the design of adaptive assistive AI systems that integrates accessibility requirements, human-centred interaction principles, and cognitively informed guidance strategies. A formative usability evaluation involving eleven visually impaired participants provides preliminary empirical evidence regarding usability, accessibility, and perceived usefulness of the proposed interaction model. The framework establishes a foundation for future research on inclusive and adaptive AI-based navigation systems in urban environments. Full article

Natural and human-made disasters can severely impair terrestrial communication infrastructures and disrupt emergency response coordination in modern smart cities. To address these challenges, this paper introduces the Weighted Average Yo-Yo-based Clustering and Routing (WAY-CR) scheme, an adaptive, secure, and energy-efficient drone-assisted solution for post-disaster network recovery and emergency response. WAY-CR integrates three main components: First, a novel WAY-based metaheuristic optimizer incorporates the concept of Yo-Yo Motion into the conventional Weighted Average Algorithm (WAA), improving the balance between exploration and exploitation during CH selection and clustering. Second, a secure communication model combines the Paillier Homomorphic Cryptosystem (PHC) with a trust evaluation model to provide end-to-end security and authenticity, ensuring that only authenticated and trustworthy drones participate in communication and routing. Third, a Trust-Aware Boltzmann Path Selection method introduces probabilistic decision-making into routing, allowing adaptive selection of secure and energy-efficient routing paths. WAY-CR formulates a multi-objective optimization model that minimizes communication cost and energy consumption while maximizing trust, link stability, and coverage. Stage 1 addresses secure intra-Ground Control Station (GCS) clustering, authentication, and trust management, whereas Stage 2 restores inter-GCS connectivity through a Secure Relay Discovery and Verification procedure based on Boltzmann Path Selection. An adaptive maintenance mechanism further supports dynamic reconfiguration in response to CH failures, mobility, or trust degradation, thereby preserving stable network performance under disaster-induced disruptions. Extensive simulation results show that WAY-CR outperforms state-of-the-art Flying Ad Hoc Network (FANET) baselines in energy efficiency, cluster stability, trust accuracy, and end-to-end packet delivery, highlighting its potential as a resilient, scalable, and secure solution for post-disaster smart-city environments. Full article

Vision–Language–Action (VLA) policies promise flexible long-horizon manipulation, but deployment under domain shift requires both reliable uncertainty estimates and a workable runtime-assurance policy. We study a model-agnostic uncertainty-calibrated safety-gating wrapper that estimates online failure risk and routes control among policy execution, pause-and-reobserve, and a fallback planner. Using a cleaned and consistently aggregated benchmark pipeline, we evaluate two long-horizon manipulation tasks in NVIDIA Isaac Sim 5.0 under lighting, texture, occlusion, sensor, and combined shifts. Relative to an ungated VLA baseline, calibrated gating improves mean shifted success from 57.5% to 77.2% and reduces aggregate expected calibration error from 0.303 to 0.100. The largest success gains occur under occlusion and combined shift, including improvements from 48.3% to 85.2% on the drawer task and from 59.4% to 87.8% on clutter sort. The results also expose a systems trade-off: an aggressive uncalibrated threshold baseline attains stronger raw success and collision metrics, but requires nearly twice as many interventions per shifted episode (21.6 vs. 11.5). The main contribution is, therefore, an empirical characterization of the reliability–intervention trade-off created by calibrated supervision, not a claim that the calibrated supervisor is universally the best terminal controller. We frame calibrated gating as a better-calibrated, lower-intervention supervisor that materially improves robustness relative to an ungated VLA while revealing the open problem of mapping calibrated risk into efficient intervention policies. Additional threshold-sensitivity, signal-diagnostic, overhead, and residual-failure analyses show that the selected operating point is meaningful but not universal: the calibrated risk threshold captures most shifted failures in retrospective logs, yet residual contacts still arise during pause and fallback states. These findings provide controlled simulation evidence for trustworthy VLA supervision under distribution shift and clarify the reliability–intervention frontier that future embodied-control systems must navigate. Full article

Reliable tabular data correction is a prerequisite for trustworthy analytics in enterprise information systems. Tabular data in such environments frequently contain formatting errors, semantic conflicts, missing values, and cross-field inconsistencies that degrade downstream analytics and machine learning performance. Rule-based methods efficiently handle structural violations but miss context-dependent errors, whereas large language models (LLMs) offer strong semantic-correction capability at inference costs prohibitive for enterprise-scale deployment. This paper formulates data error correction as a progressive decision process and proposes a complexity-aware framework with three processing stages. The first stage applies deterministic rules for low-complexity structural errors. The second stage employs a task-specialized distilled language model for medium-complexity semantic correction. The third stage performs neural probabilistic–logical reasoning on a factor graph for high-complexity cross-field errors. A learnable routing mechanism assigns each record to the appropriate stage based on a lightweight complexity score. Layer-wise conformal prediction is further introduced to construct calibrated prediction sets with coverage guarantees at each stage, together with a rejection mechanism for low-confidence corrections. The framework is evaluated on one enterprise dataset and two public benchmarks (Hospital and Flights). It improves the record-level complete repair rate by 2.1 to 3.1 percentage points over the strongest baseline (GPT-4o-Direct) and by up to 16.8 points over purely rule-based repair, while reducing average inference latency by approximately 80% relative to direct GPT-4o invocation. Ablation studies confirm the critical role of complexity-aware routing and rule-trigger features, and reliability analyses show that hierarchical conformal calibration maintains tighter coverage than single-level alternatives across varying confidence requirements. These results indicate that complexity-aware progressive routing coupled with hierarchical conformal calibration provides a practical path toward high-throughput, auditable, and reliability-controlled data cleaning suitable for enterprise deployment. Full article

The continuous advancement of autonomy and intelligence in marine shipping has made the safe and efficient navigation of unmanned surface vehicles in complex waters a major research focus. As a key link of the autonomous decision-making system for unmanned surface vehicles (USVs), local path planning needs to achieve real-time collision avoidance and motion optimization under dynamic obstacles, multiple rule constraints, and strong environmental uncertainty. In recent years, reinforcement learning has gradually become an important technical route for local path planning of USVs by virtue of its autonomous decision-making ability in high-dimensional continuous state space and adaptability to complex nonlinear problems. Combined with the evolution of the algorithm paradigm and its functional positioning in different water scenarios, this paper systematically reviews the relevant literature by examining the evolution of algorithmic paradigms; focuses on summarizing deep Q-network (DQN), Proximal Policy Optimization (PPO), Soft Actor-Critic (SAC), and Twin Delayed Deep Deterministic Policy Gradient (TD3), along with the collaborative architectures integrated with traditional planning methods such as A* and Rapidly-exploring Random Tree (RRT); and summarizes the performance characteristics, advantages, and limitations of various methods in typical scenarios. The review shows that the main bottlenecks of current research include insufficient reward mechanism design, low sample utilization efficiency, difficulty in transferring from simulation to real ships, and insufficient safety and trustworthiness verification. This paper looks forward to the future development trends from the two directions of data fusion and security enhancement in order to provide reference for related research. Full article

In today’s densely populated and technology-driven smart cities, natural and human-made disasters increasingly threaten the resilience of communication infrastructures, creating critical challenges for maintaining reliable connectivity. The failure of conventional networks during crises significantly hampers emergency response, coordination, and information dissemination. To address these challenges, this paper presents Weighted Average Algorithm-based Clustering and Routing (WAA-CR), a novel, secure, and adaptive UAV-based framework for disaster response and recovery. WAA-CR integrates three key components: shelters or Ground Control Stations (GCSs) as communication anchors and support hubs, survivable clustering and routing using a WAA-based metaheuristic optimizer, and secure and trustworthy drone communication enabled by a lightweight trust evaluation mechanism, and authentication model. The framework formulates a multi-objective optimization model that simultaneously minimizes the number of active UAVs and routing cost, while maximizing trust, communication reliability, and coverage. Cluster head (CH) election and routing decisions are guided by a composite fitness function that considers residual energy, link stability, mobility, and dynamic trust scores. Additionally, an adaptive maintenance mechanism enables dynamic reconfiguration to handle CH failures, trust degradation, or mobility-driven topology changes. Extensive simulations conducted in MATLAB R2020ademonstrate that WAA-CR significantly outperforms existing baseline FANET protocols in terms of energy efficiency, cluster stability, trust accuracy, and end-to-end delivery performance. These results validate the proposed framework’s effectiveness in building resilient, scalable, and secure UAV-based communication networks for post-disaster environments. Full article

As safety and reliability requirements continue to rise in energy storage systems and related applications, fault prognosis has become a key enabler of stable operation and proactive safety management for lithium-ion batteries. Unlike conventional fault detection and diagnosis, fault prognosis focuses on predicting the occurrence time, evolution trend, and severity of potential faults, thereby strengthening risk awareness and decision-making proactivity in battery management systems (BMSs). However, existing prognosis methods still face substantial challenges under complex operating conditions, heterogeneous data sources, and highly nonlinear degradation dynamics, resulting in limited cross-scenario generalization, unstable long-horizon prediction, and insufficient uncertainty characterization. These limitations are becoming increasingly critical as lithium-ion batteries are widely deployed in electric vehicles and large-scale energy storage systems, creating an urgent need for prognosis approaches that are more accurate, robust, and scalable. Against this backdrop, this review provides a structured and forward-looking overview of lithium-ion battery fault prognosis with three objectives: systematically summarizing representative methodological routes, clarifying key technical challenges, and identifying research priorities for intelligent prognosis enabled by pre-trained models (PTMs). Specifically, we examine recent developments across three major technical routes—model-based, signal processing-based, and artificial intelligence (AI)-based methods. Building on this synthesis, we further discuss the opportunities introduced by PTMs for battery health management and analyze the key challenges of integrating PTMs into fault prognosis. Finally, in line with the evolution of intelligent BMSs, we outline future directions for enabling efficient, reliable, and trustworthy PTM-driven applications in lithium-ion battery fault prognosis, offering forward-looking insights for next-generation intelligent battery health management. Full article

The integration of large models and multimodal foundation models into the low-altitude economy is driving a transformative shift, enabling intelligent, autonomous, and efficient operations for low-altitude vehicles (LAVs). This article provides a comprehensive analysis of the role these large models play within the smart integrated lower airspace system (SILAS), focusing on their applications across the four fundamental networks: facility, information, air route, and service. Our analysis yields several key findings, which pave the way for enhancing the application of large models in the low-altitude economy. By leveraging advanced capabilities in perception, reasoning, and interaction, large models are demonstrated to enhance critical functions such as high-precision remote sensing interpretation, robust meteorological forecasting, reliable visual localization, intelligent path planning, and collaborative multi-agent decision-making. Furthermore, we find that the integration of these models with key enabling technologies, including edge computing, sixth-generation (6G) communication networks, and integrated sensing and communication (ISAC), effectively addresses challenges related to real-time processing, resource constraints, and dynamic operational environments. Significant challenges, including sustainable operation under severe resource limitations, data security, network resilience, and system interoperability, are examined alongside potential solutions. Based on our survey, we discuss future research directions, such as the development of specialized low-altitude models, high-efficiency deployment paradigms, advanced multimodal fusion, and the establishment of trustworthy distributed intelligence frameworks. This survey offers a forward-looking perspective on this rapidly evolving field and underscores the pivotal role of large models in unlocking the full potential of the next-generation low-altitude economy. Full article

Biodiversity knowledge is fundamental to conservation planning and sustainable environmental decision-making; however, general-purpose Large Language Models (LLMs) frequently produce hallucinations when responding to biodiversity-related queries. To address this challenge, we propose BioChat, a domain-specific question-answering system that integrates a Retrieval-Augmented Generation (RAG) framework with a Re-Ranker–based retrieval and routing mechanism. The system is built upon a verified biodiversity dataset curated by the National Institute of Biological Resources (NIBR), comprising 25,593 species and approximately 970,000 structured data points. We systematically evaluate the effects of embedding selection, routing strategy, and generative model choice on factual accuracy and hallucination mitigation. Experimental results show that the proposed Re-Ranker-based routing strategy significantly improves system reliability, increasing factual accuracy from 47.9% to 71.3% and reducing hallucination rate from 34.0% to 24.4% compared with Naive RAG baseline. Among the evaluated LLMs, Qwen2-7B-Instruct achieves the highest factual accuracy, while Gemma-2-9B-Instruct demonstrates superior hallucination control. By delivering transparent, verifiable, and context-grounded biodiversity information, BioChat supports environmental education, citizen science, and evidence-based conservation policy development. This work demonstrates how trustworthy AI systems can serve as sustainability-enabling infrastructure, facilitating reliable access to biodiversity knowledge for long-term ecological conservation and informed public decision-making. Full article

With the rapid development of geological blockchains and Internet of Things-based data acquisition technologies, massive amounts of heterogeneous data are constantly emerging. However, this data is stored in a distributed manner across different organizational or business blockchains. Data sharing among multiple geological blockchains faces numerous challenges, either exposing sensitive data during verification or lacking effective authorization mechanisms. Therefore, how to achieve fine-grained access control and privacy protection across multiple blockchains has become a critical issue that must be addressed in geological data sharing. In this paper, we propose GeoCross, a cross-chain geological data sharing framework that enables fine-grained authorization management and privacy protection. First, GeoCross provides a hierarchical hybrid encryption mechanism that uses symmetric encryption for geological data protection and ciphertext-policy attribute-based encryption to enable flexible cross-chain access policies. Second, we integrate a Groth16-based zero-knowledge proof mechanism, which allows a chain to verify the existence, integrity, and accessibility of off-chain data without revealing the content. Furthermore, we introduce a Reputation-based Non-interactive Relay node Selection protocol (RNRS), which enhances the trustworthiness and fairness of cross-chain routing. Finally, we implement GeoCross in a multi-chain Hyperledger Fabric environment and evaluate its performance under real-world workloads. Results show that Groth16 verification requires only three bilinear pairings, achieving a throughput of up to 390 tps on a single chain and 1550 tps in a concurrent multi-chain environment. Even with 50% malicious nodes, the RNRS protocol still maintains a success rate of over 91%. These results demonstrate that GeoCross provides an efficient and practical solution for secure and privacy-preserving cross-chain geological data sharing. Full article

With the rapid advancement of social media and AI technologies, influencer marketing has evolved significantly. Virtual influencers have emerged as alternatives to traditional human influencers. Grounded in the Elaboration Likelihood Model (ELM), this study examines how virtual influencers’ source credibility dimensions (expertise, attractiveness, and trustworthiness) affect consumer attitudes and purchase intentions. Using the case of virtual influencer Imma, this study collected 344 valid online survey responses. The empirical results show that, along the central route, perceived product value has a significant and positive effect on purchase intention. Along the peripheral route, the trustworthiness, attractiveness, and expertise of the virtual influencer all exert significant positive effects on purchase intention. However, product involvement moderates these effects differently: for high-involvement consumers, the effects of trustworthiness and attractiveness on purchase intention are significantly strengthened, while the moderating effects on expertise and perceived value remain non-significant. This study contributes to the emerging literature on virtual influencer marketing by demonstrating how source credibility dimensions and perceived value interact with product involvement to shape consumer responses. Additionally, virtual influencers offer sustainability benefits by minimizing carbon emissions from travel and physical production inherent in traditional influencer campaigns. The findings offer practical implications for marketers: virtual influencers can effectively enhance brand exposure, but their persuasive impact varies by product involvement requiring tailored content strategies for high- versus low-involvement products. Furthermore, future research could extend this work by examining the effects of different product categories and cultural contexts on the effectiveness of virtual influencer marketing. Full article