Embodied agents need to predict the future consequences of candidate actions in order to plan effectively before execution. Existing visual dynamics models learn by reconstructing future visual states or rolling out dense latent representations, which spreads learning capacity across visually salient but planning-irrelevant content rather than the action-conditioned changes that drive manipulation outcomes. We propose CAPE, a Contrastive Action-conditioned Parallel Encoding framework that learns visual dynamics by distinguishing the future outcomes induced by different action sequences. Given an initial observation and a candidate action sequence, CAPE decodes the full future latent trajectory in a single forward pass and is trained with a Goal-Convergent Contrastive Objective that aligns predictions corresponding to the same future outcome while separating those corresponding to different outcomes. On real-world DROID and zero-shot transfer to RoboCasa, CAPE substantially outperforms prior baselines on future-state retrieval, offline action matching, and closed-loop planning, while notably reducing planning-time inference cost at long prediction horizons.

Safety is a fundamental requirement in the development of autonomous driving (AD) systems. While function offloading has demonstrated significant benefits in terms of computational efficiency and energy consumption, its application to safety-critical AD functionality introduces new challenges. In particular, offloaded service compositions incur increased and variable response times due to wireless vehicle-to-everything (V2X) communication, which directly affects the vehicle's reaction time and thus its safety guarantees. In this paper, we address this challenge by extending the definitions of Responsibility-Sensitive Safety (RSS) to explicitly account for different response times of local and offloaded AD service compositions. Based on this extension, we propose an integration into function offloading, using the RSS safety constraints for offloading decision-making and fallback mechanisms. Offloaded service compositions are only permitted if the current traffic situation remains safe under the corresponding end-to-end response time. If this condition is violated, the system performs a controlled fallback to local execution. Furthermore, we introduce an enhanced fallback strategy that includes a warm-standby phase for offloaded services, enabling faster and safer transitions from offloaded to local services. The proposed approach is integrated into our AD stack and evaluated in both simulation and the real world. Experimental results demonstrate that the proposed method improves safety compared to state-of-the-art function offloading and safety frameworks, while preserving the benefits of distributed computation when safety conditions allow.

Experience-based self-evolution is crucial for LLM agents, but existing benchmarks often assume explicit goals, stable task patterns, and clear feedback. We study a more challenging setting: low-repetition tasks with implicit rewards, where past experience is difficult to reuse and feedback is delayed, noisy, and outcome-level. We introduce \textsc{FinEvolveBench}, a temporally controlled benchmark for financial sentiment prediction that links daily news-driven predictions to future excess returns. We further propose Tree-of-Experience (ToE), a structured experience-management method that organizes, retrieves, validates, and updates agent experience. Experiments show that general-purpose experience mechanisms do not consistently outperform no-experience baselines, while ToE achieves stronger overall performance. These results highlight the importance of structured experience management for self-evolving agents in implicit-reward environments.

Reasoning Video Object Segmentation (RVOS) demands a sophisticated integration of temporal dynamics, spatial details, and linguistic reasoning to achieve precise pixel-level localization. Existing methods are limited to reasoning over fixed initial inputs and lack the capacity to actively acquire further visual evidence, which is often essential for resolving complex references in long or intricate videos. To address this, we propose \textbf{VideoSEG-O3}, the first multi-turn reinforcement learning framework for RVOS that emulates the human \textit{``coarse-to-fine''} cognitive process. It employs a \textit{multi-turn temporal-spatial chain-of-thought} to capture fine-grained details by iteratively pinpointing critical intervals and keyframes. Additionally, to enable the policy to perceive segmentation quality beyond mere text probability of \texttt{[SEG]} during the RL stage, we introduce \textit{SEG-aware logit calibration}, which integrates pixel-wise segmentation feedback directly into the token-level logits. Furthermore, we design a \textit{decoupled thinking trace} to hierarchically decompose the reasoning process into temporal, spatial, and linguistic dimensions, and construct \textbf{VTS-CoT}, a specialized cold-start dataset featuring comprehensive reasoning trajectories. The code and models will be released at https://github.com/Dmmm1997/VideoSEG-O3.

Retrieval-Augmented Generation (RAG) reduces but does not eliminate hallucination in large language models. Existing detection methods rely on flat similarity between generated answers and retrieved passages, ignoring structural relationships among evidence pieces and answer claims. We propose Evidence Graph Consistency (EGC), a framework that constructs a local evidence graph per response and computes five structural consistency measures as hallucination indicators. Evaluated on the full question answering split of RAGTruth across six LLMs (5,767 responses), EGC reveals a consistent model-family split: graph consistency features show the expected diagnostic direction for hallucinations in Llama-2 models but exhibit systematic reversal in GPT-4, GPT-3.5, and Mistral-7B. This reversal suggests qualitatively different hallucination patterns across model families and indicates that embedding-based graph consistency cannot serve as a model-independent hallucination detection signal.

Missing data is a common challenge in spatiotemporal systems, arising in applications such as air quality monitoring and urban traffic management. Traditional machine learning approaches, like recurrent and graph neural networks, rely on iterative propagation, which tends to accumulate errors over time and space. Recent diffusion-based methods mitigate error propagation but require iterative sampling and often depend on problem-agnostic Gaussian priors, limiting both efficiency and effectiveness. To address these limitations, we propose GiFlow, a Graph-Informed Flow Matching framework for spatiotemporal imputation. GiFlow replaces the typical Gaussian prior with a graph-informed prior constructed via spatiotemporal filtering of observable signals, which better aligns the source distribution to the target and thereby simplifies the generation trajectory. The flow field is parameterized by a hybrid vector field model that integrates spatial attention, temporal attention, and spatiotemporal propagation, enabling joint modeling of spatial and temporal dependencies. Extensive experiments on both synthetic and real-world datasets demonstrate that the proposed GiFlow outperforms the state-of-the-art approaches in spatiotemporal imputation. The code is available at https://github.com/zepengzhang/GiFlow.

Future AI-integrated Radio Access Networks (AI-RAN) will combine open programmability with learning-enabled xApps, rApps, and control functions that act on shared parameters and key performance indicators (KPIs). For conflict monitoring, it is not enough to know which applications are deployed; the system must also know whether the parameter--KPI dependencies assumed by runtime diagnosis remain valid under the current operating regime. This paper studies a lightweight monitoring primitive for that purpose: tracking an interpretable dependency representation from streaming telemetry events. We represent active dependencies by a Boolean matrix and use Boolean matrix multiplication to check whether recent parameter-activity and KPI-response events are consistent with the current estimate. We propose a sliding-window inference procedure that reuses the estimate when it remains consistent and recomputes it when recent observations indicate structural change. The tracker is intended as an explainable signal for conflict diagnosis and slow-loop model refresh, not as an autonomous mitigation mechanism. Experiments on controlled Boolean event streams show efficient and accurate tracking under dependency changes and Boolean observation noise.

We argue that modern generative models create structural risks for knowledge and cultural production at current, sub-AGI capability levels. We define Human Temporal Learning (HTL) as path-dependent knowledge accumulation through sustained engagement with problems over time. Generative outputs increasingly resemble HTL-intensive work in surface features, so verifying whether a given output reflects genuine human learning grows costly relative to its expected benefit. Once verification loses economic justification, evaluators reward outputs regardless of production mode, and producers who invested years of learning compete on price against outputs that cost almost nothing to generate. We call this pathway value collapse and formalize it through a costly-inspection framework. Cross-domain evidence from academic publishing, legal practice, content platforms, and software security maps onto four stages of verification erosion. Alignment success is orthogonal. Better-aligned models narrow observable gaps between human and AI outputs, making source verification harder and intensifying competitive pressure against HTL-intensive work even when individual AI outputs improve.

Symbolic regression (SR) is a class of methods that systematically explore the space of mathematical functions to discover models that accurately capture the underlying relationships in a dataset. Despite recent advances in the field, a lack of support for uncertainty quantification (UQ) limits its adoption in real-world decision processes. In regression analysis, UQ provides important information about the model reliability, which can both help to avoid overfitting by accounting for uncertainty in the data, and provide insights for decision-making. This survey is the first to clearly address this issue, with the objective of introducing essential UQ concepts and reviewing the current literature on UQ in SR, which can be broadly organized into three research directions: frequentist, Bayesian, and model selection. Despite its importance, UQ in SR is still underexplored, which motivates further research into reliable UQ methods for SR.

The critical question after a correct driving veto is not only whether a maneuver is unsafe, but whether the blocked interaction admits a lawful, auditable, and responsibility-bounded repair. Prediction and game-theoretic planners can suggest plausible cooperation, yet they do not return a proof that the repair respects hard rules, right-of-way, cost allocation, and ego fallback. We introduce CARVE, Certified Affordable Repair of Vetoed maneuvers via Envelopes, a certificate architecture for prediction-free interactive repair. Given a vetoed maneuver, CARVE constructs a finite repair lattice and emits a structured certificate recording the binding rule, selected joint repair, right-of-way-scaled cooperation envelope, responsibility-weighted cost split, and ego-only fallback. This certificate view reveals the algorithmic bottleneck: multi-owner repair induces a product lattice $M = \prod_j |\mathcal{A}_j|$. We therefore introduce CARVE-Q, a verifier-shielded quantum-AI search layer that applies quantum minimum finding only to this black-box lattice while leaving all safety authority classical. In the conservative verifier-oracle model, exact classical minimum finding requires $Θ(M)$ queries in the worst case, whereas Durr-Hoyer/Grover minimum finding uses $O(\sqrt{M})$ oracle queries with high probability. We prove verifier-shielded certificate soundness, priority non-elicitation, black-box query separation, and finite-precision reversible-oracle constructibility. We then demonstrate state-vector minimum finding on CARVE repair oracles up to 65,536 assignments and validate certificate preservation on Lanelet2-grounded INTERACTION replay with 100% right-of-way respect, 100% blame consistency, and zero priority false positives. The result is a trust-bounded quantum-AI pattern for certified autonomy: quantum proposes; CARVE certifies.

In natural images, object skeletons are used to represent geometric shapes. However, even slight variations in pose or movement can cause noticeable changes in skeleton structure, increasing the difficulty of detecting the skeleton and often resulting in discontinuous skeletons. Existing methods primarily focus on point-level skeleton point detection and overlook the importance of structural continuity in recovering complete skeletons. To address this issue, we propose Lighthouse-Skel, a topology-aware skeleton detection method via lighthouse-guided structured inference. Specifically, we introduce a dual-branch collaborative detection framework that jointly learns skeleton confidence field and structural anchors, including endpoints and junction points. The spatial distributions learned by the point branch guide the network to focus on topologically vulnerable regions, which improves the accuracy of skeleton detection. Based on the learned skeleton confidence field, we further propose a lighthouse-guided topology completion strategy, which uses detected junction points and breakpoints as lighthouses to reconnect discontinuous skeleton segments along low-cost paths, thereby improving skeleton continuity and structural integrity. Experimental results on four public datasets demonstrate that the proposed method achieves competitive detection accuracy while substantially improving skeleton connectivity and structural integrity.

This paper addresses the fusion of a pair of spatially unregistered hyperspectral image (HSI) and multispectral image (MSI) covering roughly overlapping regions. HSIs offer high spectral but low spatial resolution, while MSIs provide the opposite. The goal is to integrate their complementary information to enhance both HSI spatial resolution and MSI spectral resolution. While hyperspectral-multispectral fusion (HMF) has been widely studied, the unregistered setting remains challenging. Many existing methods focus solely on MSI super-resolution, leaving HSI unchanged. Supervised deep learning approaches were proposed for HSI super-resolution, but rely on accurate training data, which is often unavailable. Moreover, theoretical analyses largely address the co-registered case, leaving unregistered HMF poorly understood. In this work, an unsupervised framework is proposed to simultaneously super-resolve both MSI and HSI. The method integrates coupled spectral unmixing for MSI super-resolution with latent-space adversarial learning for HSI super-resolution. Theoretical guarantees on the recoverability of the super-resolution MSI and HSI are established under reasonable generative models -- providing, to our best knowledge, the first such insights for unregistered HMF. The approach is validated on semi-real and real HSI-MSI pairs across diverse conditions.

Predicting body fat can provide medical practitioners and users with essential information for preventing and diagnosing heart diseases. Hybrid machine learning models offer better performance than simple regression analysis methods by selecting relevant body measurements and capturing complex nonlinear relationships among selected features in modelling body fat prediction problems. There are, however, some disadvantages to them. Current machine learning. Modelling body fat prediction as a combinatorial single- and multi-objective optimisation problem often gets stuck in local optima. When multiple feature subsets produce similar or close predictions, avoiding local optima becomes more complex. Evolutionary feature selection has been used to solve several machine-learning-based optimisation problems. A fuzzy set theory determines appropriate levels of exploration and exploitation while managing parameterisation and computational costs. A weighted-sum body fat prediction approach was explored using evolutionary feature selection, fuzzy set theory, and machine learning algorithms, integrating contradictory metrics into a single composite goal optimised by fuzzy adaptive evolutionary feature selection. Hybrid fuzzy adaptive global learning local search universal diversity-based feature selection is applied to this single-objective feature selection-machine learning framework (FAGLSUD-based FS-ML). While using fewer features, this model achieved a more accurate and stable estimate of body fat percentage than other hybrid and state-of-the-art machine learning models. A multi-objective FAGLSUD-based FS-MLP is also proposed to analyse accuracy, stability, and dimensionality conflicts simultaneously. To make informed decisions about fat deposits in the most vital body parts and blood lipid levels, medical practitioners and users can use a well-distributed Pareto set of trade-off solutions.

This paper sets out the most general computational and econometric implications of the rigorous stochastic equilibrium theory from SELCKE (Staines (2024a)) arXiv:2312.16214. The analytical backbone is the discovery that the system converges geometrically to long-run equilibrium, at a rate given by the greater of the eigenvalue or inverse eigenvalue (from outside) closest to the unit circle and the maximum shock persistence. High-order shocks converge faster. I develop a simulation procedure to test, with asymptotic power, whether stochastic equilibrium exists for a particular model. The fundamental approximation result asserts that, whatever the order of expansion or loss function, the stochastic steady state delivers the most accurate perturbation solution. I also show that super-consistent parameter estimators $O(1/T)$ arise whenever second-order terms vanish. Besides Calvo, I study stochastic equilibrium in two alternative pricing models. Dynamics simplify considerably. I bound the time the impulse response peaks, by the maximum lag in the errors. This lends empirical support to Taylor contracts, although there are issues surrounding unit roots and the strong cost-channel. For menu costs, I demonstrate that the initial price distribution decays away super-exponentially, producing a system equivalent to Calvo with an endogenous reset probability. The impact of idiosyncratic disturbances appears as an additional wedge between actual and efficient output. Blow-up of the objective function at the boundary is proven, with the help of new distributional arguments, so the model meets existing eigenvalue existence conditions for the recursive equilibrium. Along the way, new light is shone on existing theoretical models and statistical procedures.

We introduce CausalAlpha, an open-source framework that constructs a high-frequency Geopolitical Risk (GPR) index from Telegram OSINT channels using natural language processing, and applies causal discovery methods to identify the directed causal structure between geopolitical uncertainty and financial market variables. Unlike standard sentiment indices or Granger-causality approaches, CausalAlpha employs the Peter-Clark (PC) algorithm to recover the directed acyclic graph (DAG) of causal dependencies between five category-specific GPR indicators and a set of financial variables spanning commodity prices, equity indices, and credit instruments, estimated across four DAG specifications and three significance levels with 500 block-bootstrap resamples. Two findings emerge as globally robust across all DAG specifications at alpha = 0.10: political instability and energy media coverage independently and causally precede conflict coverage, establishing conflict as the primary causal sink of geopolitical narrative escalation in real-time OSINT channels. At the strictest significance level (alpha = 0.05), conflict coverage causally precedes energy sector equity returns (delta XLE), consistent with geopolitical escalation transmitting to energy markets. A Structural VAR on the core macro panel confirms that dynamic transmission from geopolitical NLP signals to financial market prices is statistically weak at daily frequency, suggesting that geopolitical news signals operate primarily within the media narrative system. The framework is deployed as a production application on Google Cloud Run with automated data collection and index construction, representing a step toward real-time macrofinancial risk monitoring using OSINT.

A key challenge in choice modeling lies in specifying the consideration set, the subset of alternatives that individuals actually evaluate when making choices, which is unobserved (latent) to the researcher. The classical homo economicus assumption posits that individuals assess the full universal set of alternatives, a behaviorally implausible premise. Practical options include directly asking individuals, which introduces behavioral biases; treating the consideration set as a latent construct, requiring full enumeration and strong identification assumptions; or relying on ad hoc heuristics that attempt to replicate how individuals form these sets or on non-parametric methods. Recently, some researchers have used historical choices as practical consideration set, an approach made increasingly feasible by the availability of passive data sources such as smartcards, mobile phone records, and scanner data. This article provides a formal demonstration of a sufficient condition, along with Monte Carlo evidence, showing that, under a Logit data-generating process with homogeneous choice probabilities across instances, defining a practical consideration set based on historical choices yields consistent parameter estimates. The demonstration is based on a reinterpretation of the sampling-of-alternatives theorem, viewing historical choices as draws from the true consideration set, and showing that under the stated assumptions, the uniform conditioning property holds. The article concludes by discussing the practical implications of this result and potential extensions to other modeling frameworks and assumptions.

This paper presents a low-complexity motion detection method for outdoor Wi-Fi sensing based on a model-driven approach. The method exploits the structural characteristics of the phase components in channel state information (CSI) for low-multipath propagation environments, which are generally considered disadvantageous for Wi-Fi sensing, to mitigate the phase offset errors originating from wireless devices. In addition, phase averaging provides a processing gain that reduces the random noise components, including quantization and thermal noise. The theoretical basis of the method is described and its effectiveness is experimentally evaluated using Compressed Beamforming frames obtained from commercial IEEE 802.11ac devices. The experiments primarily focus wild crows flying in an outdoor orchard environment. The experimental results demonstrate that the method can detect birds even when they fly several meters away from the direct line-of-sight path between the transmitter and receiver antennas. Furthermore, the results indicated that fluctuations caused by vegetation movement were negligible when the wind speed was less than 3~m/s. The proposed approach is expected to be applicable not only to orchard monitoring but also to other outdoor Wi-Fi sensing applications in low-multipath environments.

Integrated sensing and communication (ISAC) has emerged as a key technology for 6G systems. To support the development of ISAC systems, accurate channel modeling and simulation for performance evaluation is essential. Recently, 3GPP introduced a standardized ISAC channel model and its associated calibration procedure for this purpose. However, due to the complexity of the modeling methodology and the lack of fully explicit implementation details in the 3GPP reports, different implementations may lead to inconsistent or unsynchronized simulation results. To address this issue, in this work, we implement the 3GPP ISAC channel model simulator specified in TR 38.901 and conduct a comprehensive calibration analysis. We compare the simulation results with the reference results reported by companies in 3GPP and discuss several key implementation details to provide insights into the implementation and calibration of the simulator. To facilitate reproducibility and further research, the developed simulator, together with the relevant datasets and calibration results, has been released as an open-source project on GitHub.

Uncrewed aerial vehicles (UAVs) are expected to enhance connectivity, extend network coverage, and support advanced communication services in sixth-generation (6G) cellular networks, particularly in public and civil applications. Although multi-UAV systems offer greater efficiency and cost-effectiveness than single-UAV deployments, their implementation still faces several fundamental challenges that limit their reliability, sustainability, and scalability. The limited onboard energy restricts mission duration and communication continuity. Therefore, wireless energy harvesting (EH) emerges as a promising solution to overcome this limitation. However, terrestrial energy sources experience path loss, making EH from surrounding UAVs more sustainable. Moreover, rate-splitting multiple access (RSMA) remains insufficiently explored in hierarchical UAV networks under hardware impairments (HWI) and imperfect channel state information (ICSI). This paper proposes a hierarchical ad hoc UAV network with non-linear EH and RSMA to enhance both energy and cost efficiency, where UAVs harvest energy from surrounding UAVs. For a practical scenario, we consider the effect of HWI and ICSI in our proposed system. To the best of the authors knowledge, this study is the first to investigate such a scenario in the literature. The outage probability expressions for ground Internet of things (IoT) devices, each CMU, and the overall outage probability of the proposed system are derived over Nakagami-$m$ fading channels while considering practical constraints such as HWI, ICSI, and non-linear EH. Additionally, approximate outage probability expressions are derived for high transmit power regimes. Subsequently, we formulate two optimization problems to enhance reliability and performance. Our findings indicate that the proposed system outperforms all benchmarks in terms of outage probability.

Audio reasoning requires multi-step, evidence-grounded inference over temporally dynamic and acoustically mixed signals, exceeding conventional perception tasks such as ASR or captioning. We present VISA, our submission to the Interspeech 2026 Audio Reasoning Challenge (Agent Track), evaluated via the MMAR Rubrics for correctness and reasoning quality. Under a "LALM as a Tool" paradigm, VISA strengthens large audio language models with auxiliary multi-modal evidence while avoiding heavy orchestration. The system integrates three components: multi-modal feature extraction for complementary audio and acoustic-visual clues, model-voting inference with consistency checking for stable predictions, and fine-grained category-aware routing to resolve disagreements and select rubric-aligned reasoning chains. On the official Agent Track leaderboard, VISA ranks 2nd overall with a 66.23% Rubrics score. It also achieves 77.40% Accuracy, the highest among all systems listed across both the Single Model and Agent tracks.

Video-to-audio generation has made significant progress in achieving semantic consistency and temporal alignment from silent videos. However, audio contains rich stylistic attributes such as timbre and tempo that are difficult to infer from visual and textual inputs alone. While reference audio can serve as additional conditioning, it is typically treated as a holistic signal, limiting fine-grained style control. We propose AudioIM, an attribute-aware framework that explicitly models timbre and tempo as separate control factors rather than relying on holistic prompt conditioning. Dual encoders extract complementary timbre-related and tempo-related representations, which are injected through global conditioning. A masking-based training strategy enables effective latent prompt conditioning at inference. Experiments on VGGSound show improved style similarity while preserving semantic alignment and synchronization. Audio samples are available at: https://anonymousdemo757.github.io/.

In this letter, we investigate robust beamforming design for a movable antenna (MA)-enhanced secure integrated sensing and communications (ISAC) system with imperfect eaves?dropping channel state information (CSI). To improve radar sensing performance, we formulate a radar signal-to-interference?plus-noise ratio (SINR) maximization problem by jointly opti?mizing the transmit beamforming and antenna placement while ensuring communication data security. However, the resulting op?timization problem is inherently intractable due to the nonlinea mapping from antenna positions to channel coefficients, as well as the eavesdropper (Eve) channel uncertainty. To handle these challenges, we propose a block coordinate descent (BCD)-based algorithm incorporating successive convex approximation (SCA) and fractional programming (FP) techniques. Simulation results show that our proposed algorithm exhibits fast convergence and achieves a significant improvement in the radar SINR while guaranteeing communication security.

Integrated sensing and communication (ISAC) enables sensing and communication (S&C) functionalities to share spectrum, hardware, and signal-processing resources, but the resulting inter-functionality interference creates a fundamental receiver-design challenge, particularly in uplink operation. This paper develops a rate-splitting (RS)-inspired framework for uplink near-field ISAC. The framework generalizes the sensing-centric (S-C) and communication-centric (C-C) endpoint orders of non-orthogonal multiple access (NOMA)-inspired ISAC by splitting the communication message across the sensing operation. Closed-form expressions are derived for the communication-rate (CR) and sensing-rate (SR), accounting for residual sensing interference from target-response estimation uncertainty. The achievable CR-SR rate region is characterized under sensing-matched illumination, where the proposed single-frame RS-inspired boundary contains the NOMA-inspired time-sharing region. Unlike the classical Gaussian uplink multiple access channel, where RS recovers the time-sharing dominant face, the split factor in uplink ISAC also reshapes the sensing-stage interference, allowing the RS-inspired boundary to match or strictly enlarge the S&C tradeoff. High-SNR analysis shows that, for non-aligned S&C channels, residual sensing interference changes the rate offsets but not the leading S&C slopes, whereas in the fully-aligned case it becomes slope-limiting. Using an aperture-aware near-field channel model, large-array limits are derived, showing that achievable rates remain finite as the array grows. Numerical results validate the analysis and demonstrate the benefits of the RS-inspired scheme, the impact of residual sensing interference, and the bounded large-array behaviour induced by physically consistent near-field modelling.

Reliable inspection of nanosurfaces is essential to ensure the quality of nanostructure manufacturing. Angle-resolved scatterometry provides a non-invasive inspection method that can be used in-line but often suffers from long acquisition times due to dense angular sampling. This paper addresses the data acquisition challenge by proposing an end-to-end compressed learning framework for 5-level vacancy deficiency detection in zinc oxide nanograss using ARS images. The proposed framework integrates a learnable latitude-based sampling layer with a convolutional neural network, allowing sampling and classification to be jointly optimized during training. The sampling layer exploits the physical structure of ARS patterns and learns informative latitudinal regions, which reduces the sampling search space and improves convergence. Evaluation results show that the proposed approach achieves high and stable deficiency-level classification performance under different noise conditions. Using full ARS images, the model achieves 94.2% accuracy for five-level deficiency classification and 98.6% accuracy for separating deficient from non-deficient nanosurfaces. The proposed sampling model matches full-image performance while using up to 90% fewer angular sampling points. Even when sampling points are reduced by 99.7%, the classification accuracy decreases by less than 10 percentage points. To further improve training with limited data, we also studied a GAN-based augmentation approach and used GAN-generated data for model pretraining. Augmented data resulted in fast convergence within only a few fine-tuning epochs.

Low-altitude wireless networks (LAWN) envision a reconfigurable 3D network capable of supporting mission-critical aerial operations. This paper presents a reconfigurable intelligent surface (RIS)-assisted LAWN to establish a reliable communication with an unmanned aerial vehicle (UAV) across varying wireless channel conditions and signal blockages. A low complexity stripe-based RIS phase shift optimization framework is proposed to simultaneously enhance communication reliability and provide passive sensing capability for UAV tracking under 3D mobility. Unlike high-complexity optimization approaches, the proposed method leverages the inherent structural phase-gradient of the RIS adjacent elements to significantly reduce the search space for calculating and updating the RIS configuration as the UAV moves. The analysis and simulation results demonstrate that the proposed framework outperforms conventional benchmarks in convergence speed and computational efficiency, while maintaining robust, high signal-to-noise-ratio (SNR) connectivity even in the presence of phase estimation errors and low SNR regimes. In addition, the measurement experiments using a real RIS prototype in an outdoor campus environment are performed to demonstrate the practical viability of the proposed approach.

Speech bandwidth extension (BWE) aims to reconstruct high-fidelity wideband audio from narrowband inputs. While recent approaches have made significant progress, they often struggle to reconstruct realistic high-frequency phase and harmonic structures, leading to perceptual artifacts. In this paper, we propose FSC-Net (Full-Spectrum Context Network), a parameter-efficient architecture designed to explicitly model cross-band harmonic dependencies. By integrating Fast Fourier Convolutions (FFCs) into a complex spectral mapping framework, FSC-Net expands its receptive field to the entire spectrum, capturing long-range frequency interactions effectively. To address the ill-posed nature of high-frequency generation, our novel frequency-progressive learning curriculum guides the network to reconstruct spectral details from coarse to fine. Experimental results on the VCTK and unseen EARS datasets demonstrate that FSC-Net delivers consistently strong reconstruction quality and generalization, particularly in the challenging VCTK 4 kHz-to-48 kHz task. Compared to scaled-up baselines, our model attains leading LSD and PESQ scores while maintaining a highly compact parameter footprint (1.54 M).

Driven by the emerging low-altitude economy, uncrewed aerial vehicle (UAV) swarms offer flexible integrated air-ground access and backhaul. However, providing seamless connectivity is difficult due to the interdependent dynamics of user mobility and building blockages in these 3D scenarios. These factors create rapidly shifting bottlenecks in end-to-end paths. Furthermore, the multi-dimensional nature of joint control limits the effectiveness of traditional heuristics. To address these challenges, a \textbf{\underline{M}}ulti-Scale \textbf{\underline{R}}adio \textbf{\underline{M}}ap-\textbf{\underline{G}}uided (MRMG) framework is proposed. The MRMG framework handles heterogeneous dynamics by integrating three distinct levels of radio information: global-level maps provide regional coverage insights, local-level maps capture neighborhood-scale service conditions, and link-level maps characterize high-resolution channel features. This design effectively decouples macro-movement from micro-link adaptation. To yield long-term performance improvements, A multi-agent reinforcement learning (MARL) controller learns cooperative policies for UAV movement, next-hop selection, and transmit-power control. Simulation results show that the MRMG framework not only improves network throughput but also significantly bolsters cell-edge service, nearly doubling the 5th-percentile user rate.

The extended Phaseless Rytov Approximation (xPRA) is a recently proposed device-free RF imaging technique that provides high-resolution reconstructions of the imaging region using only phaseless measurements, such as received signal strength (RSS). Because of its phaseless formulation, it can be implemented straightforwardly using existing wireless commu?nication infrastructure. It also outperforms well-known device?free phaseless RF imaging methods such as Radio Tomographic Imaging (RTI). The linear phaseless formulation used in xPRA(and RTI) makes these methods potentially useful for integrated sensing and communication (ISAC) systems in next generation wireless networks since they do not require wide bandwidths. However, so far, both xPRA and RTI have primarily been formulated in two dimensions (2D). This paper introduces a 3D extension of xPRA, which we call the extended three-dimensional phaseless Rytov approximation (x3DPRA). The novelty of our approach is that it preserves the straightforward implementation advantages of RTI and xPRA while enabling volumetric (3D) imaging. Simulation results show that x3DPRA provides good estimates of location and shape and can also reconstruct object material attenuation. We present the 3D formulation, validate it with a 2D model comparison, and report simulation results demonstrating its performance.

This letter studies variable-length finite-rate CSI feedback from a structural perspective and proposes CsiCoGen, a novel generative feedback structure with a transferable codebook mechanism without joint training. The UE maps $H_0$ into an ordered sequence of codebook indices, while the BS recursively recovers CSI from any received partial sequence of feedback indices using a shared denoising prior. This enables flexible control of feedback sequence length and per-step quantization precision through codebook size. CsiCoGen does not require jointly training a task-specific feedback encoder or codebook with the reconstructor, and the same online structure can be paired with different pretrained denoisers. In this work, we instantiate the decoder with a generative diffusion model. Simulation results on COST2100 show favorable rate-NMSE and rate-$ρ$ tradeoffs against representative baselines, with CsiCoGen reaching about -31 dB indoor NMSE and -20 dB outdoor NMSE in the high-rate regime while demonstrating scalable decoding complexity and adjustable per-step quantization precision.

Copula functions have been widely employed in wireless communication analysis to model dependence structures and evaluate system performance. However, existing studies generally express performance metrics in terms of copula dependence parameters without explicitly characterizing their admissible regions. This letter introduces the concept of copula dependence parameter regions and investigates its significance in wireless communications. Considering a two-user wireless multiple access channel (MAC) with correlated Rayleigh fading modeled by the bivariate Farlie--Gumbel--Morgenstern (FGM) copula, explicit parameter regions are derived from communication-theoretic and probabilistic perspectives using outage probability and Pearson correlation coefficient (PCC) constraints. The results show that practical communication and statistical requirements can significantly shrink the classical copula admissible interval, rendering some theoretically admissible dependence structures infeasible. Numerical examples illustrate the proposed concept and its practical implications.