Papers

Insurance fraud imposes substantial financial losses and operational inefficiencies, raising premiums and impacting trust among legitimate policyholders. Early detection at FNOL remains a persistent challenge. Existing approaches rely largely on private, text-only datasets, limiting progress on multimodal methods that integrate linguistic, behavioural, and speaker-based indicators. We introduce a synthetic multimodal framework that replicates FNOL conditions. It generates agent-customer dialogue transcripts and two-speaker audios, performs ASR and diarisation. Downstream modules combine NER, regex-based feature extraction, LLM-RAG retrieval, and speaker embeddings in a rule-based risk score to flag narrative reuse, structural inconsistencies, and cross-case voice repetition while balancing sensitivity and false positives. Dataset validation and component-level evaluations show stability and transfer potential, offering a reproducible baseline beyond text-only fraud detection.

High-resolution passive microwave imaging is important for numerical weather prediction, disaster monitoring, and oceanographic studies, but kilometer-level spatial resolution remains difficult to achieve because of aperture limitations and the high complexity of large interferometric arrays. This paper proposes a beamforming microwave interferometric radiometer (BF-MIR) for high-resolution passive microwave imaging. BF-MIR employs beamforming-capable antennas as interferometric elements in a large sparse array. The enlarged spatial-frequency sampling interval reduces the required number of elements and the cross-correlation burden, while a large aperture-to-sampling-interval ratio factor (ASRF) array design enables narrow-beam spatial filtering to suppress brightness temperature (TB) aliasing caused by spatial-frequency under sampling. In addition, beamforming enables dynamic beam steering across multiple pointing directions, thereby compensating for the limited instantaneous coverage of narrow beams. A beamforming interferometric imaging model is established, and the relationships among spatial resolution, radiometric sensitivity, and effective field of view are analyzed. An image-domain Shift-Accumulate method is further introduced to analyze aliasing, based on which an aliasing suppression strategy is developed. In addition, a three-element proof-of-concept prototype provides preliminary experimental validation of dynamic beam interferometric measurement and dynamic beam observation modes. These results indicate that BF-MIR is a promising architecture for further spaceborne high-resolution passive microwave imaging.

Long-form audio exhibits an inherent hierarchy: fine-grained events form sub-activities, which in turn constitute higher-level activities. Prior work often models these levels separately, leading to cross-level inconsistencies and requiring supervision at multiple levels. We formulate the problem as hierarchical parsing from event-level evidence: given detected event segments with class posteriors, we infer an order-consistent Act-Sub-Event parse tree. We propose Hierarchical Activity Grammar, encoding hierarchical composition and temporal-order constraints, and perform grammar-guided decoding that combines event evidence with a grammar prior. This yields a temporally grounded parse tree from which sub-activity segmentation and activity classification are derived, without requiring sub-activity or activity labels for training. Experiments on the long-form MultiAct audio dataset demonstrate improved temporal-order consistency (Edit score) and produces interpretable hierarchies.

Autonomously controlling and handling a vehicle at and beyond its stability limit is a mathematically and computationally demanding task. Prior demonstrations of automated drifting have been limited to research platforms with instantaneous torque delivery and independently actuated wheels, leaving their applicability to production vehicles with actuator latencies and mechanically coupled axles uncertain. To overcome these issues, we design a predictor to compensate for powertrain delays, develop a revised control formulation to accommodate higher actuation latencies as well as a differential coupling on the driven axle, and introduce brake-based velocity stabilization. This paper presents the controller framework, the model extensions, and real-world experimental results. We observe that our controller enables a production sports car with a combustion engine to robustly sustain circular and figure-eight drifts, limiting lateral error to 1.1 m and sideslip overshoot to 0.06 rad despite actuator delays exceeding 250 ms, while mitigating oscillations and maintaining stable path and sideslip tracking. In conclusion, our results establish that autonomous drifting is feasible on production-ready vehicles, opening pathways to advanced safety systems capable of stabilizing cars in scenarios where traditional control fails.

Reliable monitoring of hydroelectric generators requires descriptors that capture both electrical loading and electromagnetic field behavior. This work investigates operating-regime identification in the Porjus U9 10-MW Kaplan hydrogenerator using synchronized measurements from ten stator-mounted Hall probes and six rotor-current channels. Seven steady guide-vane-opening settings are considered, and each 300s record is divided into 1s windows. The resulting windows are represented by spatial Fourier descriptors of the circumferential air-gap field, probe-wise temporal flux indicators, and channel-wise RMS rotor-current features. Correlation analysis and principal component analysis are used to examine how the feature groups vary with the operating point, and Random Forest, radial-basis-function support vector classification, and multilayer perceptron models are evaluated for supervised identification of the guide-vane-opening state. The analysis shows that RMS rotor-current features mainly track the loading axis, while the magnetic-flux features reveal complementary information associated with spatial imbalance, waveform distortion, and weak low-frequency modulation. Spatial descriptors alone provide limited separability, yielding test accuracies below 27%, whereas rotor-current features alone reach about 84-85%. Combining flux and current information gives the most discriminative representation; the SVC-RBF model achieves 99.5% test accuracy and macro-F1 score. The results indicate that distributed air-gap magnetic sensing, when fused with rotor-current measurements, can support accurate and interpretable data-driven monitoring of Kaplan hydrogenerator operating regimes.

Fast charging is decisive for electric-vehicle adoption, but field chargers are deployed as one setting while the cell's true thermal state, ambient temperature, and cooling-system health are uncertain. A current that is safe for a healthy cell at room temperature can overheat the same cell when it is hot or its cooling is degraded. We formulate this as a single-setting, unknown-state safe-fast-charging problem and solve it with a margin-aware repair-before-veto controller (RACL-B). RACL-B requests an aggressive current and repairs it online to the tightest measured margin among terminal voltage, cell temperature, and negative-electrode lithium-plating overpotential, rather than committing to a fixed schedule or shutting charging down. We evaluate one deployed setting across nine conditions, spanning 10/25/40 $^\circ$C ambient temperature and 100/60/40% cooling health, in a high-fidelity Doyle–Fuller–Newman model with partially reversible lithium plating and lumped thermal coupling. Under a strict 45.0 $^\circ$C peak-temperature audit, fixed and ambient-scheduled protocols overheat in five of nine conditions because neither observes hidden cooling degradation, and rigid protective shutdown fails to deliver the charge in every condition. RACL-B safely completes all nine conditions, is 37.9% faster than the fastest fixed current safe across the whole envelope, produces the least plated lithium, and remains safe across thermal guard bands. The same margin-aware principle drives a transient-credit fault readout (CREST-B) that, on a real introduced-fault battery-pack dataset, gives the strongest learned sequence-to-global monitor for localizing cooling-fault onset under operating-condition shift. The framework provides a deployable thermal-safety guarantee for fast charging together with a margin-aware monitor for the same physical fault class.

Vehicle license plate recognition is an integral component of intelligent transportation systems. In this work, we present an embedded real-time license plate recognition system customized for developing countries. We address the challenge of handling complex, unstructured traffic scenes with diverse vehicle types while implementing the system on an embedded platform for low-cost deployment. Our method consists of license plate detection on a multi-vehicle image, followed by character recognition on the detected license plates. Both steps use lightweight convolutional neural networks to balance accuracy and efficiency. We also introduce the SL-LPR dataset of Sri Lankan road images, which contains a variety of vehicle types and traffic conditions typically seen in developing countries. On this dataset, the license plate detection and character recognition models achieved 93.6% mAP and 87.88% accuracy, respectively, and were competitive against larger models on several public datasets. To achieve real-time performance in a resource-constrained embedded environment, we applied low-bitwidth quantization using the Brevitas library and implemented FPGA acceleration for the models using the FINN framework. The end-to-end system can operate at 11.5~FPS when implemented on the Xilinx Kria KV260 platform. These results demonstrate that our system is effective for real-time license plate recognition on an embedded device, even in complex traffic scenarios. The SL-LPR dataset is available for research use at: https://github.com/sl-lpr-uom/SL-LPR.git.

In this paper, we study distributed circumnavigation of a stationary target by a heterogeneous team of agents. Each agent is modelled as a disk rather than a point mass to account for its physical dimensions. The target location is assumed to be accessible only to a small subset of agents, called leaders. The rest, called followers, therefore use only local information available from their designated out-neighbour in the interaction graph characterised by the selection of nearest neighbours. By controlling only angular speeds, we develop a distributed guidance law to circumnavigate a stationary target. The proposed guidance law works for both static and time-varying interaction graphs. Inter-agent collision avoidance is enforced through a logarithmic Barrier Lyapunov (BLF) Function, which guarantees forward invariance of the collision-free set. We show that every follower converges to circumnavigation about the same target as the leader at the end of its directed path in the interaction graph, provided the initial conditions are admissible. Numerical simulations illustrate the effectiveness of the proposed method for both static and time-varying topologies.

Prosodic emphasis varies across languages, emotions, and speaking styles, yet existing emphasis detection models are largely trained and evaluated on monolingual neutral read speech. We introduce MMEE (Multilingual Multi-Emotion Emphasis), a corpus of 10,000 professionally recorded expressive utterances (14.13 hours) across 7 languages and 34 emotion/style categories, with three-level perceptual labels (10 annotations per sample). We benchmark two state-of-the-art architectures under monolingual, cross-lingual, multilingual, cross-emotion, cross-dataset, and data-scale settings. Monolingual models show limited zero-shot transfer, degrading across typologically distant languages, while multilingual training substantially improves robustness. Models transfer robustly between high- and low-arousal emotions; bidirectional transfer between synthetic and perceptual benchmarks suggests shared prosodic structure; and performance stays robust even at smaller training scales.

Separable scheduling unleashes the deployment flexibility of mobile emergency resources by dispatching carriers and functional modules separately yet in a coordinated manner, offering a promising avenue to enhance power system resilience. However, this flexibility induces a distinct carrier-supported module routing structure, where non-self-mobile modules must be routed through compatible carrier movements. The resulting carrier-module spatio-temporal coupling makes exact and tractable optimization challenging. This letter identifies this structure and develops a novel exact bi-layer time-space network formulation as a mixed-integer linear program. The proposed formulation represents carrier and module trajectories as interacting network flows and enforces their support relations through explicit arc-level coupling. Compared with the prior logic-based model, the proposed formulation preserves exactness while improving modeling flexibility by eliminating mandatory post-arrival dwelling. Numerical studies validate its correctness and demonstrate substantial computational advantages.

Understanding human responses to autonomous vehicle (AV) behaviors is essential for socially aware interaction, which is crucial for socially compatible navigation in shared traffic environments. We characterize human driving responses in interactions with AVs as feedback laws over the coupled state space of the human driven vehicle and the AV. We model the human driver's actions using a response map, a concept based in game theory, and employ a linear representation to capture driver behaviors as a function of AV behaviors, based on empirical data from a driving simulator study. Our results show that 1) human driver acceleration behavior can be captured using response maps, and 2) human driver responses differ significantly with respect to AV behaviors of yielding, non-yielding, and responsive to the human driver.

Lighting is the dominant energy load in indoor farming, yet most deployed systems still rely on fixed rule-based or schedule-based control. We present LightFARM, a predictive lighting control framework that couples crop illumination with battery-free sensing for more energy-efficient indoor farming. LightFARM combines finite-horizon predictive control with compact models of photosynthesis, thermal dynamics, and sensor energy state. The controller adjusts lighting intensity to balance photosynthetic benefit, electrical power consumption, thermal safety, and sensing-energy feasibility. A key design feature is that the same light-emitting diode (LED) fixtures serve both as the photosynthetic light source for crops and as a controllable energy source for self-powered sensor nodes. We implement LightFARM in a real indoor basil cultivation system and evaluate it through two independent 12-day cultivation trials. Compared with a conventional rule-based baseline, LightFARM reduces lighting energy consumption by approximately 41% and improves energy productivity from 36.1 to 52.9 $\mathrm{g\,kWh^{-1}}$ and from 41.1 to 60.2 $\mathrm{g\,kWh^{-1}}$ ($\approx 46.5\%$ on average). These results suggest that energy-cooperative predictive lighting control is a promising approach to improving indoor farming efficiency under practical resource constraints, while explicitly accounting for the trade-off between energy savings and crop yield.

Reliable, low-latency communication is critical for real-time monitoring and control in modern Smart Grids (SGs). The emergence of 5G networks, with enhanced reliability, significantly lower latency, and native support for massive machine-type communication, offers strong potential to enable advanced grid applications such as state estimation (SE) and fault detection. While existing studies investigate 5G for SG use cases, most rely on simulations or analytical models; experimental validation using real hardware and SG data remains limited. This paper fills this gap by presenting a fully experimental validation of real-time SE over a commercial 5G network using a 5G-based multi-node testbed built with Raspberry Pi (RPi)-based SG nodes and a Typhoon Hardware-in-the-Loop (HIL) real-time simulator. We first characterize 5G communication performance using simulated SG data under varying reporting rates and deployment environments by evaluating Key Performance Indicators (KPIs) such as end-to-end delay, jitter, and frame loss. Experimental results show that the worst-case mean delay observed for the 5G is approximately 6.5x lower than that of our previous LTE cat-M study at the corresponding reporting rate. We then stream real-time voltage, current, and phase-angle measurements-generated by an IEEE 4-node feeder model in Typhoon HIL simulator-to a remote Phasor Data Concentrator (PDC) for SE and fault detection. Results demonstrate that 5G-enabled measurements support accurate SE under both steady-state and dynamic load variations. Furthermore, fault-detection experiments confirm reliable and prompt fault detection, with detection delays as low as 0.80 s.

Cross-platform deployment of offensive comment detection for Chinese social media suffers performance degradation. The paper proposes a dual-threshold hard mining method to address this. First, the clean-Chinese-base RoBERTa is finetuned on COLD to establish a binary baseline for fair comparison. Second, a three-class fine-labeled test set covering Weibo, Xiaohongshu, Tieba, and Zhihu is constructed, domain distances from the source are quantified using Jaccard and Proxy-A Distance, as well as the degradation bottleneck of the baseline under domain shift is systematically revealed. Herein, a dual threshold hard example mining strategy is proposed. High- and low-confidence error-prone samples are filtered from unlabeled corpora by prediction confidence. The model is secondarily finetuned under implicit contexts with merely a small set of manually labeled hard examples, realizing low-cost cross-platform domain adaptation. Experiments reveal significant performance gains of the optimized model across four platforms.

Silicon photonics enables integration of optical components using standard semiconductor processes, greatly improving data communication bandwidth and energy efficiency. However, photonics integrated circuits (PICs) face unique security challenges, such as counterfeit or tampering threats, that conventional electronic security methods do not address. We propose a novel hardware fingerprinting technique that embeds two dimensional photonic crystal patterns into the density control filler regions of a PIC. Each PhC pattern is designed to resonate a specific visible to near infrared wavelengths, producing a distinctive optical signature (based on wavelength, polarization, and incident angle) for each device. Finite difference time domain (FDTD) simulation using ANSYS Lumerical is employed to optimize nanostructure dimensions and spacing so that each device's reflection/absorption spectrum contains unique narrowband peaks. No extra fabrication steps or materials are required beyond standard lithography, keeping costs low. The embedded nanostructures have sub-50nm precision, making forgery extremely difficult. Our method yields a high resolution, scalable fingerprint for silicon photonic chips, enabling cost-effective device authentication and improved supply chain security.

Photonic Integrated Circuits (PICs) are advancing high-performance computing, data centers, and sensing, yet three-dimensional (3D) PICs introduce critical thermal management challenges due to high-density bonding and heterogeneous materials. Traditional methods like thermal microscopes and in-package sensors yield sparse data, limiting full thermal profile visibility. This paper presents a dual-method solution combining an AI-driven thermal modeling framework with a design-based heuristic approach. The AI method integrates sparse sensor data with design layer and density information to predict multilayer temperature variations, while the heuristic approach uses localized material properties, design layout, component geometries, and sensor coordinates to refine thermal estimations in specific regions. A 2D thermal map of a 3D PIC is generated by interpolating sensor data and adjusting for local thermal resistivity using comparative analysis between design regions. The heuristic method complements the AI model, improving estimation accuracy without extensive training data. Together, these methods offer a scalable, accurate solution for real-time thermal mapping and design-time simulation, enabling reliable thermal management in next-generation 3D photonic systems.

This paper presents a novel priority random access (PRA) non-orthogonal multiple access assisted ALOHA, called PRA-NA, to provide access priority for machine-type devices (MTDs) with different delay requirements (i.e., delay-sensitive and delay-tolerant). We first introduce a received power level model that incorporates imperfect channel state information and imperfect successive interference cancellation to study the impact of practical non-ideal channel conditions. Two PRA strategies including fixed PRA-NA (FPRA-NA) and adaptive PRA-NA (APRA-NA) are then designed to reduce the average access delay of delay-sensitive MTDs in heterogeneous massive machine-type communications. Subsequently, the throughputs of both the FPRA-NA and APRA-NA strategies are analyzed to demonstrate their effectiveness. Moreover, to improve the energy efficiency of random access, we introduce an enhanced user barring algorithm (EUBA) to carry out power control. It is shown that our proposed EUBA can not only alleviate the user overload problem, but also reduce the average transmit power of MTDs. By extending it to the proposed PRA-NA schemes, we demonstrate via extensive simulation results that the random access performances in terms of throughput, access delay, and energy efficiency can be significantly improved over the conventional NOMA-ALOHA.

Physical or operational constraints often impose communications limitations on autonomous agents. Such limitations complicate monitoring or multiagent coordination. Even when strong communications are absent, some information may still be available. The remainder of the relevant agent state may be reconstructed via estimation. The actions taken by an agent are a potential source of information – as the agent interacts with the environment, these actions may be observed even in the absence of explicit communication. We investigate using actions to estimate the state of an agent, using reinforcement learning to develop policies which make the estimation problem more tractable. Policy observability is encouraged through the training reward and is analyzed using simulation of the trained agent. In an aircraft tracking problem a policy with enhanced observability is found that has minimal impact on nominal task performance.

In this paper, we present an experimental evaluation study of the Alternating Direction Method of Multipliers (ADMM), which is a widely used technique in the distributed optimization of power distribution networks. The focus of this study is on how real 5G communication performance affects ADMM in a fully experimental platform that features commercial 5G connectivity and real-time control. The ADMM-based Distributed Optimal Power Flow (DOPF) problem is solved using the IEEE 123-bus unbalanced distribution feeder subdivided into five areas, each managed by a local controller implemented on a Raspberry Pi. To mitigate the impact of the communication network variability, we propose a delay threshold-based mechanism that yields a 7.75% reduction in convergence time compared to a no-threshold baseline. We also devised a policy to dynamically update the threshold value based on communication and computation conditions, achieving a 26.42% reduction in the convergence time compared with the static optimal threshold. These results demonstrate the potential of adaptive, communication-aware control strategies for real-world Smart Grid (SG) deployments.

This letter presents the design and operation of an electronically reconfigurable pinching antenna (E-pinching antenna) and examines its capability to establish controllable millimeter-wave links that can circumvent blockages. The antenna consists of a low-loss rectangular dielectric waveguide that leaks energy through modular varactor-loaded elements to form tunable radiation points. A copper reflector ensures unidirectional radiation, thereby boosting forward-link efficiency and spatial selectivity. Full-wave simulations demonstrate that the radiated power and transmission coefficient to a receiving patch antenna can be dynamically tuned by adjusting the varactor capacitances. A multi-user scenario is investigated by activating two radiation modules along the waveguide to serve spatially separated receiving antennas isolated by a metallic partition (blockage). Simulation results confirm the capability to establish links to both LoS and NLoS users with minimal propagation losses. The proposed architecture enables a scalable and electronically controllable distributed antenna platform for reconfigurable wireless systems with enhanced blockage mitigation.

We characterize a notion of system comparison, termed as $(T_e,\gamma,\delta)$-similarity, for non-deterministic nonlinear systems. Building on a similar notion recently proposed for stable linear systems, the proposed notion characterizes the dissimilarity between the outputs, measured using the $L_2$ norm, of two nonlinear dynamical systems in terms of their inputs and disturbances. By establishing a relationship between $(T_e,\gamma,\delta)$-similarity and differential dissipativity, we establish equivalence between $(T_e,\gamma,\delta)$-similarity of nonlinear systems and the $(T_e,\gamma,\delta)$-similarity of their differential dynamics. We characterize the $(T_e,\gamma,\delta)$-similarity for nonlinear systems as a Linear Matrix Inequality feasibility problem and also provide necessary and sufficient conditions for solving this feasibility problem. We demonstrate the utility of the proposed notion through its use in two applications: (i) robust hierarchical control applied to a planar aircraft and (ii) the improvement (or design) of abstract models applied to the Moore-Greitzer model and an electronic circuit.

Large-aperture wave receivers can contain a large number of candidate sensor locations, antenna ports, or measurement blocks, while hardware and processing constraints allow only a subset to be activated. In this paper, receiver selection is formulated as a finite-SNR modal-recoverability problem, where the selected subset is required to support reliable recovery in every direction of a prescribed modal subspace. However, large trace, log-det, or codebook-distance values alone do not ensure that every prescribed modal direction is resolved. Specifically, the proposed framework consists of three nested recoverability criteria for individual modal degrees, a joint target subspace, and target modes in the presence of active non-target modes. The third criterion, the Schur-complement information floor, provides an exact worst-direction posterior-error interpretation. We further show that stricter recoverability criteria require at least as many active receiver accesses and derive tests that identify reliability targets that remain unattainable even when all available accesses are activated. Next, we specialize the framework to finite spherical-wave sampling and compare greedy receiver-selection rules. Numerical results demonstrate that global log-det is generally more access-efficient at moderate reliability floors, whereas Schur-based selection is more effective at stringent floors. While this paper is motivated by holographic and XL-MIMO receivers, the framework can be applied to general sampled wave systems.

We address the problem of inferring a directed network from nodal measurements generated by linear diffusion dynamics on the sought graph. Observations are modeled as the outputs of a graph convolutional filter, i.e., a polynomial (with unknown coefficients) of a local diffusion graph-shift operator encoding the latent graph topology, excited with an ensemble of independent graph signals with arbitrarily-correlated nodal components. Unlike prior efforts that considered undirected graphs and white signal excitations, here the graph-shift operator and the observations' covariance matrix are not simultaneously diagonalizable. In this challenging context, we first rely on measurements of the output signals along with prior statistical information on the inputs to identify the diffusion filter. Such system identification problem involves solving a system of quadratic matrix equations, which we show is identifiable under spectral-diversity assumptions on the input covariances. For algorithmic purposes we recast it as a smooth quadratic minimization subject to Stiefel manifold constraints. Subsequent identification of the network topology given the graph filter estimate boils down to finding a sparse and structurally admissible shift that commutes with the given filter, thus, forcing the latter to be a polynomial in the sought graph-shift operator. A joint graph filter and topology identification algorithm is also proposed, which alternates between the aforementioned steps in a mutually reinforcing fashion to offer improved sample complexity. Numerical tests corroborate the effectiveness of the proposed algorithms in recovering synthetic digraphs and real-data case studies, and illustrate their potential utility on urban mobility analyses as well as portfolio optimization.

3D Scene Graph Generation (3DSGG) represents 3D scenes as structured object–relation–object graphs for spatial understanding. In observer-centric spatial perception, the same scene may be expressed under different local observer frames while its structure remains unchanged. However, existing models typically assume a fixed scene-aligned reference frame and may produce semantically inconsistent predictions when the scene is re-expressed in another observer frame. We attribute this failure to the heterogeneous frame dependency of relational predicates. Directional predicates such as $\textit{left}$, $\textit{front}$, $\textit{right}$, and $\textit{behind}$ are $\textbf{Observer-Dependent Relations}$, whereas most contact, support, and semantic predicates, such as $\textit{standing on}$ and $\textit{attached to}$, are approximately $\textbf{Observer-Independent Relations}$. Conventional models do not distinguish these frame responses, leading to degraded relation prediction under observer-frame reorientation. We introduce $\textbf{Observer-Aware Relations (OAR)}$, which combines observer-aware geometric encoding and relation specialization, supported by frame-stable object encoding, for unified multi-label predicate prediction. Experiments on 3DSSG show that OAR consistently outperforms baselines across controlled observer-frame reorientations without training-time frame-reorientation augmentation, while remaining competitive on the standard benchmark. The project page is available at https://oar-predicate.github.io/.

We study a quantum autoencoder (QAE) for compression-driven anomaly detection in brain MRI data. The approach leverages angle encoding to map image patches into quantum states, followed by a variational encoder-decoder architecture trained to discard information via auxiliary trash qubits. Anomaly scores reflect the degree to which inputs resist compression relative to normal data, with higher scores corresponding to deviations from the learned normal manifold. Evaluated on publicly available brain MRI DICOM datasets, the method achieves a slice-level ROC-AUC of approximately 0.95 and a patch-level ROC-AUC of approximately 0.813, outperforming classical autoencoder and PCA baselines. Analysis of the learned parameters reveals a pronounced encoder-decoder asymmetry, where effective anomaly detection arises from structured information compression within the encoder rather than increased parameter magnitude or decoder expressivity. This results in a controlled compression-reconstruction trade-off with a clear operating regime that supports principled threshold selection. Qualitative evaluation further shows that the QAE produces spatially localized anomaly heatmaps aligned with tumorous regions. The results, supported by promising baseline performances, demonstrate that quantum autoencoders provide an interpretable and controllable mechanism for anomaly detection based on incompressibility with respect to a learned latent representation. This work highlights the potential of quantum autoencoders as a principled tool for studying compression dynamics in quantum machine learning, with promising implications for decision support in medical imaging workflows.

The primary goal of Remote Sensing Image Change Captioning (RSICC) is to automatically generate descriptions of changes between remote sensing images captured at different time points. Existing models still rely on a single autoregressive generation paradigm, which tends to prioritize learning easily generated vocabulary over capturing discriminative differences between images. To address this, we reframe the training paradigm and propose a novel Difference Feature Modeling (DFM) framework. Specifically, we introduce a Text-guided Gated Contrastive Loss (TGCL) to guide the vision encoder to extract critical features from a text-modal perspective. Additionally, we incorporate a pre-trained Change Detection model to transfer stable change detection knowledge. In order to further enhance the representation, we design a Joint Feature Modeling (JFM) module to achieve the fusion of multi-scale difference representations, thereby capturing comprehensive spatiotemporal variations between multi-temporal images. Extensive experiments on multiple datasets demonstrate the effectiveness of our approach.

Multi-agent large language model (LLM) systems often rely on verifier and critic agents to suppress hallucinations, but verification is delayed. During this delay, false claims can propagate through the agent network. We model this process as delayed consensus on a graph with grounded corrector nodes. Spectral decomposition by the grounded Laplacian yields a closed-form stability threshold for the verification dose: correction that is too strong or too delayed can turn consensus into oscillation. The most unstable regime occurs when the communication and verification delays coincide; for delay two, the threshold is the inverse golden ratio. The same framework gives a supermodular placement objective and a greedy (1-1/e)-approximation rule for assigning a limited corrector budget to influential nodes. Experiments across five open models confirm the predicted dose-delay oscillations. By contrast, grounded factual answering makes truth an absorbing boundary and eliminates the effect, suggesting that the instability is specific to signed-belief tasks while grounded verification remains stabilizing

Deep learning has shown significant potential in medical image analysis, particularly for disease detection using MRI scans. Accurate and early diagnosis of brain tumors remains challenging due to the complexity of brain structures and reliance on manual interpretation. This work presents an automated deep learning-based approach for brain tumor detection from MRI images using Convolutional Neural Networks and Residual Networks. Transfer learning is applied with two pretrained architectures, ResNet18 and ResNet50, to classify MRI scans into tumor and non-tumor categories. Experiments are conducted on a dataset of 3,929 brain MRI images, evaluating the impact of model depth and fine-tuning strategies. The results show that ResNet18 achieves a higher accuracy of 97% compared to 96% for ResNet50, demonstrating better generalization on limited medical data. The proposed framework enables fast, accurate, and cost-effective brain tumor detection, supporting early diagnosis and clinical decision-making.

Modern high-performance mobile electronics impose extreme constraints on thermal management, and traditional cooling methods often fail to meet requirements for power density, form factor, and durability. Jet impingement cooling offers a compelling solution but is typically hindered by the need for bulky ancillary hardware. Here, we demonstrate that compact arrays of reduced-scale electroaerodynamic (EAD) plasma actuators, which are silent, solid-state devices with no moving parts, can be used for direct jet impingement cooling of electronics. The main contribution is the first rigorous experimental demonstration and system-level validation of multi-stage, ducted electroaerodynamic jet arrays as a compact, fan-replacement impingement cooling solution for mobile electronics. We characterize the performance of both single- and multi-stage ducted actuators, including thermographic analysis of heat transfer coefficients and spatial cooling profiles. We also quantify the relationship between actuator stage count and cooling efficiency, showing that increasing the number of ion acceleration stages enhances jet velocity and heat transfer performance at a reduced efficiency. The actuators are then assembled into an array and directly compared to a conventional fan with similar coverage area, showing competitive performance at a fraction of the volume, weight, and power. Finally, we integrate the array onto a commercial edge AI system and show that thermal regulation during extended inference workloads matches that of a stock fan, without any moving mechanical components or noise. These results confirm that multi-stage EAD jet arrays are not only viable but advantageous for thermal management in mobile and high-performance systems, paving the way toward silent and miniaturized solid-state cooling solutions.

Neural audio autoencoders have become a core component of compression, feature extraction, and generation. However, while existing systems support variable bitrate, the vast majority of models still operate at a fixed latent frame-rate, allocating equal temporal budget to regions with very different information density, which can result in unnecessarily long sequences. We introduce Elastic Time, a dynamic frame-rate bottleneck that converts fixed-frame-rate autoencoders to dynamic ones. Our method learns a lightweight latent predictor used to decide which frames can be skipped and later reconstructed, enabling efficient greedy boundary selection at inference. Experiments show our method enables deployment-time rate control while improving efficiency-quality tradeoffs relative to baselines. Overall, we provide a flexible mechanism for adjusting temporal resolution in audio autoencoders, potentially facilitating more efficient downstream modeling for generation and long-context tasks.