Papers

A defining feature of 6G networks is that performance depends not only on the quantity of available resources (e.g., spectrum, antennas, cache memory, compute, and fronthaul bandwidth) but also on their fungibility, i.e., the ability of one resource to substitute for another under changing conditions. We argue that the fungibility landscape of a distributed 6G system is governed by two coupled decision scales: micro decisions made locally by agents and macro outcomes that emerge at the network level. Existing distributed-optimization approaches largely conflate these scales. To address this gap, we develop an agent-based-modeling (ABM) framework that separates macro and micro decisions through three operator-controllable macro choices, three micro hyperparameters, and three structural metrics. We establish six key results: (i) a two-timescale decomposition theorem, (ii) a structural-metric basis theorem, (iii) a macro–micro design rule with closed-form factorization of the emergent breakdown threshold, (iv) a fungibility–resilience monotonicity proposition, (v) a connectivity–substitutability duality theorem, and (vi) a multi-application generalization proposition. Numerical results visualize the macro fungibility landscape and the micro decision-sensitivity region for a representative 6G deployment.

This paper presents a two-dimensional method-of-moments (MoM) solver for electromagnetic scattering from infinitely long perfectly electrically conducting (PEC) cylinders. Both TMz and TEz polarizations are considered. Starting from the scalar Helmholtz equation, the electric field integral equation (EFIE) is derived for TMz scattering and the magnetic field integral equation (MFIE) is derived for TEz scattering. The induced surface current on the PEC boundary is expanded using pulse basis functions, and the boundary integral equations are discretized using point matching at the segment centers. Circular cylinders with radii $R = {\lambda}$ and $R = 2{\lambda}$ are used as validation cases because analytical series solutions are available. The MoM-computed surface currents, total near fields, scattered near fields, and field-error distributions are compared against the analytical solutions. After validation, the same solver is applied to a square PEC cylinder, for which no simple closed-form analytical solution is used. The results show strong agreement between the MoM and analytical circular-cylinder solutions and demonstrate the geometry-dependent scattering behavior of the square cylinder.

Cardiac magnetic resonance imaging (CMR) routinely records structured acquisition metadata, yet most CMR foundation models rely primarily on image-only pre-training and leave this naturally available source of weak semantic supervision largely underexplored. We propose MetaCLIP-CMR, a metadata-driven framework based on Contrastive Language–Image Pre-training (CLIP), which converts imaging modality, anatomical view, scanner vendor, field strength, and scanner model into textual supervision for CMR representation learning. The pretrained image encoder is evaluated on imaging modality classification, cine view classification, and cardiac segmentation. MetaCLIP-CMR achieves 86.8% modality accuracy and 86.5% cine view accuracy, clearly outperforming ImageNet and masked reconstruction initialisations. For downstream cardiac segmentation, MetaCLIP-CMR consistently obtains the highest Dice score across the evaluated ACDC and M&Ms cine short-axis (SAX) settings under both full-data and 20% fine-tuning regimes. Compared with recent image-focused large-scale CMR pre-training models, MetaCLIP-CMR achieves comparable ACDC segmentation performance, while requiring less than 1% of their pre-training image scale. These results suggest that metadata learning offers a natural and easy-to-use strategy for transforming routinely recorded acquisition information into effective supervision for foundation-level CMR representation learning, highlighting the promise of metadata-driven multimodal pre-training.

Machine learning for underwater acoustics is constrained by the scarcity of publicly available labeled datasets. In contrast to air-acoustic domains, where large benchmarks enable rapid model development, underwater datasets are typically small and limited in acoustic diversity, restricting robust model training and cross-domain generalization. To help address this gap, we introduce a curated underwater audio dataset derived from an open-source maritime sound archive. The dataset contains over one thousand labeled audio segments across eight biologically and mechanically relevant acoustic classes, providing an additional resource for training models in data-limited underwater environments. Additionally, we establish a lightweight Convolutional Neural Network (CNN) baseline and propose a margin-enhanced loss with feature alignment to mitigate class confusion arising from data imbalance, acoustic similarity, and cross-domain mismatch. While the baseline achieves 96.35% in-domain accuracy, evaluation on ShipsEar reveals substantial domain shift; the proposed feature alignment improve zero-shot ship detection by 42.60%, demonstrating stronger robustness under distribution mismatch. We further release a transparent curation pipeline and reproducible benchmark to support future research on imbalance mitigation, domain adaptation, and data-efficient underwater acoustic classification.

Decarbonisation is raising the electrical load on mainline alternating-current railway feeders that were not designed for sustained, simultaneous high-power demand. When several trains accelerate together on a shared feeder, the contact-line voltage can fall far enough to trigger rolling-stock current limitation or feeder protection, eroding capacity and reliability. Preventing this in real time requires a quantity conventional operation does not expose: a localised, continuously updated estimate of the traction power available to each train given the live network state. A railway power-flow model, with trains represented under a voltage-dependent automatic current-limitation characteristic, shows that the minimum network voltage is governed by the product of power and distance rather than by power alone, yielding a distance-dependent single-train power envelope. This envelope does not add up when several trains share a feeder, so a conservative pairwise screen is generalised to a solver-free multi-train estimate: a calibrated shared-path voltage model returning the minimum section voltage and the per-train available power for any number of trains. Calibration uses two short offline solver runs, one fixing the self-impedance and one the inter-train coupling through a separation-dependent factor. Its current-limitation behaviour follows EN 50388-1, and on matched multi-train cases the estimate tracks the full power flow to within about nine per cent on average across two-, three-, and four-train cases, improving as more trains share the feeder, while its online cost scales with the number of trains rather than the network size.

We propose and analyze a communication-centric reconfigurable intelligent surface (RIS)-assisted integrated sensing and communication (ISAC) system, in which a monostatic radar simultaneously senses a moving target and serves a user equipment (UE) over Nakagami-m fading. We design a dual-function phase-modulated continuous-wave (PMCW) waveform that embeds the data stream directly into the radar pulse train: each pulse carries one full maximum-length sequence whose polarity is flipped by a binary phase-shift keying data symbol, so that the same emission preserves the sharp range autocorrelation required for sensing while conveying one bit per pulse to the UE. We further propose a communication-centric RIS phase configuration that co-phases each element onto the direct radar-to-UE path, yielding a coherent superposition at the UE and a received-power gain that scales with the square of the number of elements. We show that from the radar's perspective, however, the same surface behaves as an uncontrolled scatterer, since the resulting reflection paths are mis-phased and do not benefit from array combining. We derive a closed-form approximation for the average UE bit error rate based on a moment-matched Gamma approximation, and we show that the same waveform still forms a usable range-Doppler map for sensing. Monte-Carlo simulations corroborate the analytical results.

Different synthetic aperture radar (SAR) sensors vary significantly in resolution, polarization modes, and frequency bands, making it difficult to directly apply existing models to newly launched SAR satellites. These new systems require large amounts of labeled data for model retraining, but collecting sufficient data in a short time is often infeasible. To address this contradiction, this paper proposes a data generation and transfer framework, integrating a stable diffusion model with attention distillation, that leverages historical SAR data to synthesize training data tailored to the unique characteristics of new SAR systems. Specifically, we fine-tune the low-rank adaptation (LoRA) modules within the multimodal diffusion transformer (MM-DiT) architecture to enable class-controllable SAR image generation guided by textual prompts. To ensure that the generated images reflect the statistical properties and imaging characteristics of the target SAR system, we further introduce an attention distillation mechanism that transfers sensor-specific features, such as spatial texture, speckle distribution, and structural patterns, from real target-domain data to the generative model. Extensive experiments on multi-class aircraft target datasets from two real spaceborne SAR systems demonstrate the effectiveness of the proposed approach in alleviating data scarcity and supporting cross-sensor remote sensing applications.

The performance of rate-splitting multiple access (RSMA) can be severely affected by imperfect successive interference cancellation (SIC) in practical wireless systems. This paper investigates a downlink pinching antenna system (PASS)-assisted RSMA network under imperfect SIC, where residual common-stream interference is explicitly incorporated into private-stream decoding. To improve user fairness, a max-min rate optimization problem is formulated through the joint design of antenna activation, common-rate allocation, and power allocation. The resulting mixed-integer non-convex problem is addressed using a two-stage framework that combines greedy channel-aware antenna activation with successive convex approximation (SCA)-based resource allocation. Numerical results demonstrate the effectiveness of the proposed framework in improving fairness under imperfect SIC.

Synthetic aperture radar (SAR) data augmentation is important for improving the generalization of data-driven SAR interpretation models, yet practical augmentation workflows are often hindered by heterogeneous dataset formats, task-dependent metadata requirements, diverse generation methods, and weak validation of generated samples. This paper presents the SAR Augmentation and Generation Agent (SAGA), a schema-grounded and benefit-aware agent framework for task-oriented SAR data generation and augmentation. Given a natural-language request and heterogeneous SAR inputs, SAGA extracts observable dataset facts, validates executable dataset schemas, selects feasible augmentation strategies through validator-constrained planning, and compiles the selected strategy into an auditable augmentation workflow. Generated data are further assessed by quality, distribution, SAR-artifact, duplicate, leakage, and optional downstream-task evaluators to support evidence-qualified augmentation claims. By separating semantic proposal from deterministic validation and execution, SAGA improves the reliability and reproducibility of SAR augmentation decisions. Experiments on controlled agentic benchmarks and downstream SAR interpretation tasks show that SAGA improves schema grounding, skill planning, invalid-sample rejection, and downstream augmentation utility compared with rule-based, LLM-only, ReAct-style, and fixed-augmentation baselines.

As antenna arrays grow, near-field effects become non-negligible in large-scale MIMO, making accurate low-overhead channel acquisition crucial in both far-field and near-field regimes. Deep-learning-based frequency-domain channel extrapolation can reduce pilot overhead, but existing extrapolators generalize poorly to unseen distances and environments, especially across near-field and far-field channels. We propose a physically interpretable framework to unify generalization across both regimes. Our key insight is that angular profiles are regime-dependent, while delay profiles share a sparsity structure that can be aligned. Based on this, we develop a physics-guided disentanglement and alignment pipeline with multi-cluster decoupling, angle-delay feature disentanglement, and delay-domain alignment, enabling the model to learn distribution-stable delay features while reusing heterogeneous angular features. We further design a unified near/far-field DL extrapolator (UNiFi-DLE) and detail its dataset preparation, training, and inference. Simulations and sim-to-real experiments show that UNiFi-DLE generalizes robustly to unseen near-field and far-field scenarios and consistently outperforms state-of-the-art methods.

While modern ASR systems achieve low error rates on high-resource benchmarks, such performance often overestimates real-world robustness. Existing evaluations address challenges in isolation, lacking a unified benchmark for domain terminology, age variation, dialects, accents, and low-resource languages, particularly across the Middle East and Southeast Asia, representing over one billion under-evaluated speakers. To address this gap, we introduce GigaSpeechBench, a comprehensive multilingual and multidimensional in-the-wild ASR & AST benchmark comprising 680 hours of human-annotated speech. It features five modules: (1) 12 low-resource Middle Eastern and Southeast Asian languages, plus challenging Japanese and Korean; (2) 6 Chinese dialects; (3) 6 English accents; (4) dense terminology across 12 vertical domains for Chinese and English; and (5) older adult and child speech. We further provide human-annotated Chinese and English translations for 11 languages to support AST evaluation. Extensive evaluations of leading foundation models and commercial APIs reveal significant performance degradation in these challenging settings, exposing critical evaluation blind spots.

The adaptive moment estimation algorithm, known as Adam, is widely used in modern machine learning, owing to its low per-iteration complexity and strong empirical performance. Despite its prevalent use, the theoretical foundation of Adam remains largely unexplored for time-varying and nonstationary systems. In fact, the existing theoretical analyses of Adam-type algorithms are primarily concerned with time-invariant model parameters and explicitly or implicitly rely on independent and identically distributed (i.i.d.) data assumptions, under which the learning taskcan be formulated as minimizing a fixed expected objective with a static minimizer. However, such assumptions are often violated in time-varying and nonstationary systems, thereby calling for a theoretical investigation beyond the conventional yet idealized i.i.d. setting. The main objective of this paper is to solve this challenging problem by establishing a general theory of Adam for time-varying and nonstationary stochastic systems. We will introduce some new techniques for analyzing the products of nonstationary and dependent random matrices induced by Adam's coupled first- and second-moment recursions, and will construct a new stochastic Lyapunov function that blends these two moment dynamics. Under a stochastic excitation condition that allows nonstationary and dependent data, we will derive both parameter tracking and output prediction error bounds explicitly, quantifying the effects of stepsize, first- and second-momentum parameters, gradient noise and parameter drift. These bounds not only provide guarantees for Adam performance, but also provide guidelines for hyperparameter selection. Experiments on both synthetic and real-world data validate our theory and design guidelines.

Wireless foundation models (WFMs) have recently emerged as a promising paradigm for learning multiple channel state information (CSI) acquisition tasks. However, unlike natural language tokens governed by statistical co-occurrence, wireless channels are generated by electromagnetic propagation laws, and current WFM training is constrained by limited data scale, narrow distribution coverage dominated by simulations, and a pronounced sim-to-real gap. As a result, simply scaling model parameters and CSI samples does not necessarily yield robust and generalizable models. In this paper, we advocate enabling physics equivariance as a principled and explainable inductive bias for WFMs. Specifically, we focus on a universal propagation property for electromagnetic waves, termed wave equivariance: when the input CSI is modulated along time-frequency-space dimensions, the output channel response should exhibit the corresponding transformation. Empirical studies show that the vanilla-WFM fails to reliably acquire such equivariance even with a large number of model parameters and training samples. To address this, we design the physics-intrinsic WFM (phys-WFM) with wave equivariance, which explicitly aligns model behaviors with an interpretable wave propagation structure. Results demonstrate that the proposed design effectively captures wave equivariance and substantially improves robustness and generalization to unseen environments under distribution shift, offering a physics-grounded and testable route toward explainable wireless foundation models.

We study the capacity of an additive white Gaussian noise (AWGN) channel followed by a subtractive dithered uniform quantizer. Under the Schuchman conditions and with negligible overload probability, the system admits an additive-noise representation in which the effective noise is the sum of Gaussian and uniform components. Capacity bounds are derived for this model where inputs are subject to an average-power constraint as well as a peak-amplitude constraint, where the latter accounts for the limited quantizer dynamic range. Specifically, a computable lower bound is obtained based on the entropy power inequality (EPI), using the maximum-entropy input under the above constraints. Tighter numerical lower bounds are derived using discrete input constellations with finite mass points. Finally, an upper bound is obtained by exploiting the maximum-entropy property of the Gaussian distribution for a given variance. Numerical results show that, for a K-level quantizer, discrete constellations with K mass points already achieve near-optimal rates among the tested families. Moreover, our upper bound is close to the lower bounds in the moderate signal-to-noise ratio (SNR) regime; thus it provides a simple capacity approximation in this regime.

Power delivery – including high-to-low voltage conversion, complex power distribution across heterogeneously integrated chiplets, and efficient interconnect allocation – remains a critical bottleneck in high-performance computing (HPC) systems. Existing vertical power delivery (VPD) solutions are estimated to achieve less than 70% system-wide end-to-end power delivery efficiency, defined from platform input power to delivered on-chip load power, with substantial energy lost as heat before reaching on-chip point-of-loads (POLs). In the absence of systematic design methodologies, evaluating power quality, exploring architectural alternatives, and optimizing performance rely on computationally prohibitive simulations, resulting in suboptimal designs. This paper introduces an end-to-end scalable power delivery framework for HPC systems, including distributed VPD (DVPD) architecture, DVPD design optimization methodology, and analytical models. The framework leverages substrate-embedded GaN power switches together with arrays of unit inductors and capacitors tailored for HPC applications. Multi-stage power conversion schemes (48V-to-1V, 48V-to-24V-to-1V, and 48V-to-12V-to-1V) are explored, with system-wide voltage drops and power losses evaluated under steady-state conditions. Design specifications for passive and active devices are formulated to meet next-generation efficiency targets. For the 48V-to-1V case, the proposed DVPD approach achieves 84% system-wide efficiency while occupying 54% of the area beneath the load system, with efficiency increasing to 87.6% at 75% area utilization across a 1–50~kW load range. Furthermore, steady-state voltage drops peak at 2.7% and transient drops at 9% (without decoupling capacitors), demonstrating the viability of DVPD for future wafer-scale HPC platforms.

Variational quantum reinforcement learning offers a compact policy class for building-energy control, but it inherits a deployment weakness shared by learned controllers: when the thermal model is locally wrong, a policy that appears safe on the model can violate occupant comfort in the real building. Guarantees that depend on noisy quantum read-out are also insufficient for safety-critical control. We address this gap with Q-DASC, Discrepancy-Attributed Safe Quantum Control. Q-DASC wraps a variational-quantum-circuit (VQC) policy with a certified classical safety layer that discovers misspecified operating regimes with false-discovery-rate control, repairs their local thermal gains with shrinkage, projects the proposed quantum schedule onto the repaired comfort-feasible set, and attributes residual violations to policy error, model error, or physical limits. Because the final certificate is produced by classical projection, comfort feasibility is invariant to finite-shot and depolarizing read-out noise. On real BOPTEST building emulators across three buildings, two localized misspecifications, and three seeds, Q-DASC reduces average comfort violation from 26.0% for the raw VQC controller and 55.3% for a model-trusting scheduler to 0.02%, matching a clairvoyant oracle, and remains at 0.24% under NISQ read-out noise. A repair-aware VQC variant reaches 0.00% violation and reduces projection intervention, while the default Q-DASC keeps lower energy and stronger observational-data behavior. The same wrapper transfers to EnergyPlus heating and cooling benchmarks and to real hospital air-handling-unit data. These results establish a safety-efficiency frontier for deploying quantum policies in physics-constrained control.

This paper develops a continuum theory of exit-and-join coalition dynamics in nonatomic cooperative games. We extend the Aumann-Shapley value and the Aumann-Drèze value to coalition structures in which each coalition is treated as a restricted nonatomic game, yielding a marginal-contribution-based payoff density that governs incentives for agents to remain in, exit, or join coalitions. We derive deterministic mean-field dynamics from decentralized switching rules and show that payoff-difference switching recovers replicator dynamics as a special case. We characterize exit-and-join equilibrium by the absence of profitable positive-mass deviations and prove its equivalence with stationarity of the induced mass dynamics under incentive-compatible and strictly payoff-responsive switching rates. For mass-based cooperative games, we construct a Lyapunov function and establish global convergence under strict concavity. We further show that the equilibrium is equivalent to a Wardrop equilibrium of an induced nonatomic population game and admits a variational inequality formulation. The framework is extended to incorporate switching costs and endogenous coalition acceptance rules, leading to constrained equilibria characterized by quasi-variational inequalities. The proposed theory unifies cooperative value allocation, noncooperative coalition mobility, mean-field dynamics, evolutionary game theory, and population games within a common framework for analyzing coalition formation and adaptation in large-scale multi-agent systems.

Feasibility-restoration layers turn learned, market-based, or optimizer-generated decisions into actions satisfying hard constraints in systems such as power grids. Yet a repair is only as trustworthy as its constraint model: line parameters, sensitivities, ratings, and topology can be locally wrong, so a decision certified feasible under the nominal model may violate the deployed system. We identify this false safety as a dominant failure mode of model-trusting repair and propose Trust-Calibrated Certified Repair (TCR). TCR treats repair as trust calibration and answers four questions in one pipeline: where the physical model is wrong, discovered from measurements with false-discovery control; how much each constraint should be trusted, set by test-gated shrinkage and uncertainty-proportional security margins; what least-cost intervention restores feasibility, computed by a certified repair program; and why the cost was paid, attributed to genuine congestion versus avoidable model error through dual prices. On a physically grounded dynamic-line-rating benchmark whose true ratings follow IEEE 738 under real weather, TCR reaches 98% true-network feasibility, within two points of a clairvoyant oracle, at lower-than-naive cost and with perfect localization. Model-trusting repair, robust margins, and chance-constrained tightening leave substantial feasibility or cost gaps. The same method transfers unchanged to transmission redispatch over PGLib-OPF networks and distribution voltage regulation on the IEEE 33-bus feeder. Across all three task families, TCR gives the strongest deployable feasibility-cost frontier under localized physical-model misspecification. Calibrating trust in the constraint model is the missing ingredient for reliable AI-assisted engineering decisions.

This article surveys spatial-domain-enhanced Physical-layer Authentication (PLA), with Dual-polarized Antennas (DPA), Massive Multiple-Input Multiple-Output (MIMO), and Reconfigurable Intelligent Surfaces (RIS) as the primary focus. With the rapid growth of wireless deployments, authentication mechanisms face stringent requirements for high security, low overhead, and low latency. PLA offers lightweight identity verification by exploiting physical-layer characteristics. However, the effectiveness of PLA critically depends on how physical observations are constructed and validated under wireless channels. Unlike existing surveys that mainly organize PLA by authentication modality, feature source, and evaluation metrics, this work emphasizes the connection between spatial-domain enhancement mechanisms, the resulting feature representation, and the authentication procedure. We review how DPA, Massive MIMO, and RIS reshape PLA feature representation, and we summarize newly introduced security threats along with representative defense strategies. Case studies further illustrate the practical impact, such as representative detection-probability trends across Signal-to-Noise Ratio regimes and quantitative comparisons among representative schemes. Finally, we outline promising future opportunities enabled by Dynamic Metasurface Antennas, Extra-large MIMO, and spatial configuration with artificial intelligence.

Specific emitter identification (SEI) determines which known emitter a received signal originates from, while specific emitter verification (SEV) determines whether the received signal genuinely comes from its claimed emitter. In this paper, we consider the effect of wireless fading channels on SEI and SEV. When the Rician $K$-factor varies, the resulting distribution shift induced by the channel degrades both identification and verification performance. To address this issue, we first theoretically prove that the coefficient of variation of the signal envelope is strictly monotonic with respect to the Rician $K$-factor. Motivated by this property, we propose an envelope-guided adaptive feature modulation (EAFM) identifier for SEI and an EAFM with Mahalanobis distance metric learning (EAFM-MD) verifier for SEV. Specifically, the proposed EAFM identifier adopts a dual-branch neural network to extract device-oriented features from the IQ-domain input and channel-conditioning features from the normalized signal envelope, and adaptively modulates the former via feature-wise linear modulation. Then, we extend the EAFM identifier to an EAFM-MD verifier. The device-fingerprint library is constructed by storing the feature centroid and covariance for each enrolled device, along with the within-device Mahalanobis distances of training signals. For verification, the Mahalanobis distance between the extracted test features and each stored centroid is computed using the stored covariance matrix, and the minimum distance is compared to the corresponding device threshold to make a decision. Finally, numerical results show that the proposed EAFM identifier improves cross-channel identification performance, while the proposed EAFM-MD verifier achieves superior detection performance against unknown spoofing attacks.

In next-generation wireless networks, the growing density of devices and limited spectrum resources pose severe jamming challenges to fragile legitimate communication links in the wireless electromagnetic environment. Crucially, when jamming overlaps with pilot and data symbols in both time and frequency domains, it inflicts a severe bottleneck on receiver-side joint estimation and detection. Existing schemes often lack an effective framework to combat such jamming contamination, thereby failing to guarantee reliable transmission. To address this issue, we propose a Brownian bridge diffusion-based joint channel estimation and data detection framework (BBD-JCED) for jamming-resilient receivers. Specifically, the proposed framework comprises two core modules: the first extracts jamming features in the short-time Fourier transform (STFT) domain and suppresses jamming samples, thereby improving the signal-to-jamming-plus-noise ratio (SJNR) of the received signal; the second introduces a Brownian bridge diffusion (BBD) process to model the evolution of the suppressed signal and the encoded bits in the presence of channel estimation errors, thereby enabling enhanced joint channel estimation and data detection. To alleviate the computational burden of the BBD process in the second module, we further derive a fast ordinary differential equation (ODE) solver that enables its low-complexity iterative evolution. Finally, we design a multi-module training algorithm to improve the data recovery capability of the proposed framework. Simulation results demonstrate that the proposed framework achieves superior bit recovery performance compared with baseline schemes while maintaining a lower number of model parameters and competitive computational complexity.

Continuous-recording surveillance systems face a storage problem that codec tuning alone cannot fully solve: even at aggressive CRF settings, a static-camera scene spends most of its bits re-encoding a background that has not changed. We present BLUE, a pre-encode compositor that exploits this structure by maintaining a persistent seed frame of the background and substituting background pixels with seed pixels before the encoder runs. The encoder then emits near-free SKIP macroblocks for the frozen background, while live pixels in foreground regions are carried unchanged at full quality. We evaluate BLUE on all 308 annotated short subclips from the VIRAT Ground Surveillance Release 2.0 dataset using a six-point CRF sweep with both x264 and x265. At CRF 28, BLUE reduces file size by a mean of 34.6% (x264) / 39.4% (x265) on 95.8% / 99.4% of clips respectively. Foreground-region PSNR, computed only over VIRAT object-annotation bounding boxes, is preserved or improved on 60.7% of clips (+0.36 dB mean, +5.48 dB maximum). Full-frame perceptual quality (VMAF) drops by a median of 6.75-8.59 points; we quantify and disclose this trade-off explicitly. A lightweight deployment gate measuring the compositor's own VMAF on a 2-second prefix identifies the 40% of clips where even full-frame quality degradation is near-imperceptible (Delta VMAF <= -2.9), enabling a selective-activation strategy that retains both the storage benefit and acceptable perceptual fidelity.

Non-autoregressive automatic speech recognition (ASR) enables parallel decoding, but many refinement-based methods begin from random, fully masked, or fixed-length token sequences, requiring multiple iterations to reconstruct the complete transcript. We instead formulate ASR decoding as a variable-length edit refinement of a greedy connectionist temporal classification (CTC) hypothesis. An acoustic-conditioned Edit Flow decoder operates directly on the collapsed CTC hypothesis, predicting insertion, deletion, and substitution operations in parallel. The Edit Flow decoder is jointly trained with a CTC model using a continuous-time discrete diffusion loss. During inference, we find that just two edit steps yield substantial Word Error Rate (WER) reductions, and classifier-free guidance (CFG) further enhances recognition quality by focusing the model on audio features. We also constrain edit proposals using CTC confidence to improve accuracy. Finally, ablation studies validate our design choices, while decoder pretraining and pretrained encoder integration yield significant additional performance gains.

Leveraging large-scale weakly supervised datasets is crucial to train robust end-to-end automatic speech recognition (ASR) models. However, such datasets often contain noisy labels and lack domain specificity, limiting their effectiveness. To address these issues and make better use of weakly supervised datasets, we propose a novel training approach incorporating data filtering and selection. Our approach consists of three steps: pretraining on the entire dataset, continued pretraining on a filtered subset based on character error rate (CER), and fine-tuning on a small number of acoustically similar samples to the target domain, selected from the filtered subset. In experiments with a 90,000-hour weakly supervised Japanese dataset, the proposed filtering and selection methods synergistically reduced CER by up to 6.4% and 4.0%, respectively, even though these steps reused training samples already used in the first pretraining step.

This paper addresses the harmonic instability problem of the virtual-admittance (VA)-based grid-forming control. It is revealed that the intra-loop coupling among the VA control, the inner-loop current control, and the voltage feedforward control results in an \(s^2\)-term in the equivalent output impedance of the inverter, which induces a negative-resistance property in the harmonic range. It is worth highlighting that this negative resistance is independent of the control delay. Consequently, this harmonic instability mechanism is fundamentally different from the extensively investigated cases in the literature, which are induced by the digital control delay of inverters. Then, a simple passivity-oriented damping control is proposed to mitigate the negative resistance arising from the intra-loop coupling. The method fully retains the well-established current controller and voltage feedforward, and does not require grid impedance information. Finally, experimental tests verify the theoretical findings and the effectiveness of the damping method.

Causally linking disease-related factors to image-derived biomarkers provides a powerful pathway to understanding disease mechanisms. Despite growing interest in applying causal artificial intelligence (AI) approaches for this task, these methods still need to be adapted for complex medical images, and especially, neuroimaging. However, the lack of ground-truth data presents a barrier to development. To bridge this gap, we developed and tested a method for generating synthetic neuroimages, which adhere to a user-specified causal structure describing the non-image to image variable relationships, permitting the creation of ground-truth neuroimaging datasets. In the simulated T1-weighted magnetic resonance images, anatomical variability is modeled by sampling from a subspace estimated from real data and deforming a template image to create unique simulated subjects. Causal relationships are encoded via precise volumetric changes of any region-of-interest without unwanted global artifacts. We achieved relative volume errors of 0.3-2.66% for the targeted regions-of-interest and demonstrate their statistically significant causal relationships, while maintaining mean absolute errors for non-target brain regions between 0.034-0.397ml. An initial evaluation of causal discovery methods exposes their limited ability to suppress spurious connections, highlighting the need for image-appropriate methods. Our framework is the first to enable the generation of realistic synthetic 3D neuroimages with explicit causal control that can serve as the missing ground-truth data necessary for the objective benchmarking and development of causal AI methods.

The Karhunen–Loève transform (KLT) diagonalizes the covariance of a second-order process and is optimal for mean-square truncation. Which classical transform it reduces to is governed by the symmetry commutant of the covariance: when the kernel commutes with a group action, the KLT eigenfunctions are the irreducible representation functions of that group, recovering the Fourier, cosine, Mellin, and spherical-harmonic systems. We study the inverse question. Given a covariance $R$ and a finite-dimensional space of candidate generators, the generator nearest to commuting with $R$, the minimizer of $\delta(A,R)=\|[R,A]\|_F/(\|R\|_F\|A\|_F)$, is the smallest-eigenvalue solution of a double-commutator eigenvalue problem $\mathrm{ad}_R^2(A^\ast)=\lambda A^\ast$, a Hermitian generalized eigenvalue problem of size the number of generators, independent of dimension. The framework recovers hidden transforms as well as classical ones: a variational characterization turns the existence of a commuting generator into a spectral condition, and a tridiagonal commutant-uniqueness result yields the prolate spheroidal, cosine, and discrete orthogonal-polynomial bases as exact recoveries, with matrix-valued extensions, and produces a continuum of transforms interpolating between and beyond the classical families. When symmetry is approximate, the coding penalty of the symmetry-adapted blockwise transform equals the multi-information among the sectors, an exact threshold between the fixed and data-driven transforms. We further give a graph-automorphism characterization of permutation structure, a sequential deflation for non-Abelian symmetry, and stability bounds under estimation error. As an application, the KLT of a two-paradigm covariance is synthesized from its two known generators, without forming the mixed covariance, reaching the full-data transform's compaction from few observations.

Accurate and timely channel state information (CSI) is essential for next-generation wireless systems, yet existing works treat CSI compression and CSI prediction as separate problems, both in academia and in current 3GPP studies. Consequently, channel aging remains insufficiently addressed within standardized CSI feedback pipelines. In this article, we propose a unified compression-prediction framework that integrates Contrastive Predictive Coding (CPC) directly into the 3GPP-compliant CSI compression architecture. Instead of predicting high-dimensional CSI matrices, our approach forecasts future latent representations and jointly optimizes reconstruction fidelity and temporal predictive coherence via a combined 1-SGCS and InfoNCE objective. This design enables temporal representation learning without increasing feedback overhead. We present two variants: CPC-before-Compression, which performs autoregressive modeling on encoded features prior to quantization, and CPC-after-Compression, which shifts temporal modeling to the base-station to reduce the complexity of users' devices. Evaluations on 3GPP-compliant datasets from Nokia, Oppo, and CATT show that CPC-before-Compression achieves over 90% reconstruction accuracy with 32x lower decoder GFLOPs than the 3GPP baseline, while CPC-after-Compression preserves an identical encoder footprint and the same 64-bit feedback overhead. By unifying compression and prediction within a standardized pipeline, the proposed framework provides an age-aware, computationally efficient CSI feedback solution. The source code is publicly available at: https://github.com/AhmedRadwan02/cpc-3gpp

Early detection of dementia through speech analysis offers a non-invasive screening alternative, but capturing both acoustic and linguistic biomarkers remains challenging. We propose a multimodal framework leveraging Whisper for dual-purpose extraction: acoustic representations from encoder outputs and transcripts via automatic speech recognition (ASR). For the acoustic pathway, temporal networks with attention pooling aggregate variable-length sequences into fixed-dimensional embeddings. For the linguistic pathway, we prompt a large language model (LLM) to extract interpretable features spanning lexical diversity, syntactic complexity, semantic coherence, and discourse patterns. A gated fusion network integrates both modalities. On ADReSS and ADReSSo, our method achieves F1-scores of 89.47% and 90.14%, demonstrating effective integration of acoustic and LLM-augmented linguistic features. Ablation shows that multimodal fusion consistently outperforms either modality alone.

This letter studies the budgeted scheduling of stealthy false data-injection (FDI) attacks against state estimators in cyber-physical systems. Existing event-based attack schedulers require full knowledge of the plant model and assume the residual distribution is exactly Gaussian – assumptions that fail for real-world CPS sensor streams whose residuals are heavy-tailed and whose dynamics are unknown to the adversary. We propose a model-free attack-scheduler that replaces the parametric Gaussian threshold with the empirical quantile of a learned sequence autoencoder residual, calibrated from measurements alone without any plant matrices. We prove that the realized attack rate converges almost surely to the target budget under stationary ergodic residuals. Experiments on two synthetic systems and a real heavy-duty truck dataset show that the proposed scheduler tracks the budget to within 1-2% while also preserving the residual magnitude, guaranteeing stealthiness against any residual-based detector. Comparing with the model-based baseline – granted the true plant and innovation covariance – mis-realizes the budget by up to 8.96% under heavy-tailed residual distribution, causing the attacker to achieve only 1.37x system degradation when 1.84x is intended.