This paper develops a functional theory for multi-target detection, where a compactly supported signal is recovered from a single noisy observation containing many unknown translations of the signal. Our formulation allows continuous, off-grid translations and correlated stationary Gaussian process noise, extending beyond the discrete, grid-aligned, white-noise models common in prior work. We analyze two uninitialized recovery algorithms based on autocorrelation analysis; in particular, both algorithms first estimate the signal's bispectrum via a debiased third-order empirical autocorrelation. The signal is then recovered from the estimated bispectrum using either a functional frequency marching scheme or a Kotlarski-type deconvolution formula. For both algorithms, we prove non-asymptotic recovery guarantees for compactly supported signals without bandlimiting assumptions. The resulting error bounds depend on the smoothness of the signal and the accuracy of bispectrum estimation, with the latter governed by the noise characteristics and the number of signal occurrences. Numerical experiments validate our theory and demonstrate accurate recovery in low-SNR regimes.

Microwave linear analog computers (MiLACs) have recently emerged to enable high-performance and efficient beamforming in the analog domain. In this paper, we introduce a dual-functionality framework for MiLAC-aided transceivers. Beyond analog-domain precoding/combining (active beamforming), a MiLAC and its antenna array can simultaneously act as a reconfigurable intelligent surface (RIS) (passive beamforming). This allows the MiLAC to execute beamforming for transmission/reception while reflecting external incident signals. We provide an optimal reconfiguration strategy for this dual-functional MiLAC, and characterize the fundamental limits on the trade-off between active and passive rate, namely the capacity region bounds and the sum-rate capacity.

Satellite communications are envisioned as a key enabler for ubiquitous coverage in future 6G networks, yet the broadcast nature renders them vulnerable to eavesdropping, especially given the long-distance transmissions and associated high uncertainties. In this paper, we propose the physical layer security enhancement for multi-beam satellite communications with the assistance of an aerial reconfigurable intelligent surface (ARIS). Considering the high dynamics and uncertainties of channels, we characterize the channel distribution with moment-based ambiguity sets. Accordingly, a distributionally robust secrecy rate optimization is formulated through joint design of transmit and reflection beamforming. We then introduce a conditional value-at-risk-based reformulation to convert the probabilistic constraints into deterministic forms. An alternating optimization framework is subsequently employed to iteratively update the transmit and reflective beamforming vectors until convergence. Simulation results demonstrate that the proposed distributionally robust scheme significantly enhances secrecy performance, and maintains reliable performance across various channel error distributions.

Cell-free massive multi-input-multi-output (CFmMIMO) communication networks aim to provide uniform quality of service by distributing access points (APs) across a coverage area. In user-centric variants, each user equipment (UE) can choose a cluster of APs with the best channel conditions (e.g., the closest APs) for accessing service. This approach eliminates the notion of cells with dedicated regions and APs, as found in cellular mMIMO communication networks. Estimating uplink channels between UEs and APs is a crucial step in CFmMIMO communication networks; however, existing channel estimation (CE) approaches typically originate from mMIMO systems without considering the unique properties of CFmMIMO communication networks. For instance, shorter AP-UE distances in CFmMIMO systems result in Rician channel models with prominent line of sight (LoS) components between APs and UEs, motivating cooperation between APs for improved performance. In this paper, we propose a cooperative minimum-mean-squared-error (MMSE)-based uplink CE approach where APs share their linearly compressed signals as fused signals with other APs in the same cluster. The proposed approach is optimal, i.e., its performance is equivalent to that of the centralized CE approach, where APs share their uncompressed raw signals. Notably, this optimality is achieved in one shot; that is, given the required correlation matrices, the optimal fusion filters and estimators are derived non-iteratively. Consequently, the proposed approach guarantees lower communication overhead for cooperative CE compared to the centralized approach. Numerical experiments corroborate the superior performance of the proposed cooperative CE approaches in terms of CE accuracy and convergence rate.

For high-throughput applications such as ultra-high-definition video streaming and immersive extended-reality, perceptual quality rather than bit-level accuracy defines the primary performance criterion and provides a more informative and spectrally efficient objective than strict bitwise reconstruction. This is particularly relevant in millimeter-wave (mmWave) and sub-Terahertz (sub-THz) systems, where path loss, short channel coherence times and phase noise introduce severe fluctuations that degrade link spectral efficiency. We propose an extension to conventional Adaptive Modulation and Coding (AMC) framework that incorporates perceptual quality awareness into link adaptation. In this framework, the decision metric is a Perceptual Quality Indicator (PQI) derived from the Structural Similarity Index Measure (SSIM). The receiver employs a Denoising Convolutional Neural Network (DnCNN) denoiser to enhance post-decoding image quality before feedback estimation. The resulting perceptual metric replaces the standard Channel Quality Indicator (CQI) in the AMC loop, enabling adaptation to maximize spectral efficiency while satisfying a perceptual-fidelity constraint. Experiments on a 5G-compliant mmWave testbed demonstrate up to a twofold gain in spectral efficiency while maintaining perceptual fidelity, underscoring the potential of perception-optimized link adaptation.

Large language models (LLMs) are increasingly used to automate power-system analysis, but many utilities and energy-research labs require on-premise serving for confidentiality, regulatory, reproducibility, and cost reasons. This makes the reliability of open-weight models a deployment issue. We show that first-pass failures in power-system code generation are dominated not by reasoning alone, but by structured API-knowledge boundary errors: hallucinated function names, misused parameters, and mishandled result tables in versioned simulation libraries. We introduce PowerCodeBench, an execution-validated benchmark generator that pairs natural-language operator queries with pandapower code and numerical ground truth; an L0-L3 documentation-driven probing procedure that measures per-model API knowledge profiles; and a boundary-aware intervention that combines query-side API demand estimation with targeted proactive documentation injection and routed reactive correction. On a 2,000-task frozen release, we evaluate ten open-weight LLMs (1.5B-480B parameters) and four commercial mid-tier APIs. The intervention improves every evaluated open-weight model of at least 7B parameters and every commercial API by 32 to 56 accuracy points. Open-weight models in the 70B-120B range match the commercial mid-tier accuracy range, while Llama-3.1-405B and Qwen3-Coder-480B lead the panel. The targeted prompts preserve the full-context accuracy ceiling while using 41% of the prompt-token cost. The result is an accuracy-side, deployment-time path toward reliable on-premise LLM assistance for grid-analysis workflows without fine-tuning or cloud inference.

Conversational automatic speech recognition in Hungarian is constrained by the limited amount of publicly available dialogue-style training data. The BEA-Dialogue corpus addresses this need, but its strictly speaker-disjoint train/dev/eval split reduces the usable material to only 85 hours. In this paper, we introduce BEA-Dialogue+, an expanded version of the corpus that relaxes the split criterion for experimenters and dialogue partners while preserving complete separation of the primary speakers. This results in 200 hours of transcribed natural conversations and enables a controlled study of the trade-off between additional training data and speaker overlap across the splits. We evaluate several Whisper- and FastConformer-based models on both corpus versions, including Serialized Output Training (SOT)-based fine-tuning for dialogue transcription. Our results show that the larger corpus is more challenging for models without fine-tuning, whereas SOT-based adaptation yields consistent improvements in WER, CER, cpWER, and cpCER. Overall, BEA-Dialogue+ provides a substantially larger yet still demanding benchmark for Hungarian dialogue ASR, and a practical resource for training and evaluating dialogue transcription systems.

Reasoning-based robotic policies using large language and vision-language models achieve strong semantic planning capabilities but mostly suffer from a high inference latency that limits practical real-time deployment. In this work, we observe that robotic reasoning workloads contain substantial temporal redundancy, where consecutive observations frequently produce identical actions and subgoals. Based on this insight, we present REIS, a human cognition inspired robotic decision-making framework that minimizes unnecessary reasoning while preserving semantic adaptability. REIS combines lightweight scene gating, KV-steered affordance routing, and deliberative reasoning to accelerate robotic control under embodied constraints. Experiments on ALFRED, and real-world robotic tasks demonstrate that REIS significantly suppresses reasoning overhead while maintaining competitive task performance.

Ladder logic translation is an important problem in industrial automation because without it, it is difficult to switch Programmable Logic Controller (PLC) vendors. The prevailing translation problem highlights mismatched programming environments, incompatible ladder logic constructs, limitations in terms of differences in the semantic expressiveness of the vendor formalisms and integrated black-box proprietary engineering tools which are exemplified in our example case; Rockwell to Siemens PLC code translation. This work presents a mathematical formulation of the problem, the detailed architecture of a solution which supports XML extraction, structural normalization, constrained generative function (LLM), and system integration via the TIA Portal Openness API as rigorously engineered pipeline for automated translation of Rockwell Ladder Programs to Siemens S7 ladder programs. Finally, we present results that show that the translations retain high semantic consistency across instruction categories.

Image Scanning Microscopy (ISM) is a fluorescence imaging technique that combines detector-array acquisition and computational reconstruction to achieve the theoretical resolution of an ideal confocal microscope, i.e., one operating with an infinitesimally small pinhole, while maintaining high signal-to-noise ratio. Among the reconstruction methods for obtaining the super-resolved image, multi-image deconvolution (MID) and its extension aimed at preserving the optical sectioning capability of confocal microscopy, known as super-resolution sectioning ISM (s$^2$ISM), are among the most widely used approaches. Both methods rely on Richardson--Lucy-type iterative schemes, whose semi-convergent behavior requires early stopping and often leads to noise amplification and reconstruction artifacts. In this work, we introduce a self-tuning explicit regularization framework for both MID and s$^2$ISM reconstruction. Within a Bayesian maximum a posteriori formulation, we combine a multi-frame Poisson data fidelity term with explicit regularization, considering $\ell_1$ and smoothed total variation penalties as representative examples. We further develop an automatic and ground-truth-free strategy for regularization parameter selection by adapting the residual whiteness principle to the multi-frame Poisson setting and introducing a spectral high-pass extension tailored to s$^2$ISM. The resulting framework enables stable reconstructions without empirical stopping rules. To demonstrate the proposed framework, we consider first-order optimization schemes based on proximal gradient and mirror descent methods with adaptive backtracking strategies. Experiments on simulated and real fluorescence ISM datasets demonstrate improved reconstruction stability and image quality with respect to unregularized approaches, while enabling robust super-resolution and optical sectioning in low-photon conditions.

The electrification of on-road fleet logistics promises improved air quality, lower noise emissions, major climate benefits, increased energy flexibility through the use of locally generated electricity and reduced dependence on imported fuels. However, battery electric vehicles can introduce operational planning challenges not present with internal combustion engine vehicles, including heterogeneous charging speeds, exposure to volatile electricity prices, and scarcity in infrastructure. Managing these complexities requires solutions that balance cost efficiency and robustness, supported by sector coupling between transport and electricity systems. This paper reviews the current state of digital systems for operational decision-making in electric fleet management through a grey literature analysis, drawing on practitioner-oriented sources such as industry reports, company documentation, and technical blogs that reflect real-world practices and developments. We identify key trends and gaps, providing insights to guide future research and development.

This paper proposes an integrated sensing and communication (ISAC)-enabled grant-free uplink framework based on artificial-path delay modulation. A grant-free user equipment (g-UE) conveys uplink information by modulating the delay of a controllable artificial path derived from the scheduled downlink waveform. In contrast to conventional superposition-based schemes with successive interference cancellation, the proposed method enables uplink-downlink coexistence in the delay-sensing domain. By introducing a single weak artificial path confined within the cyclic prefix (CP), the g-UE allows the access point (AP) to decode uplink symbols from CSI perturbations while causing only limited degradation to the scheduled user equipment (s-UE) in the downlink. To support reliable finite-alphabet delay detection under unknown path gain and off-grid leakage, we develop a baseline delay calibration procedure and a normalized matched-filter detector. Results show that reflection power determines the reliability trade-off between the g-UE and the s-UE, whereas the delay step mainly controls the g-UE reliability-efficiency trade-off with little additional impact on the downlink s-UE. Even with an artificial path 15 dB weaker than the scheduled downlink signal, the g-UE achieves lower BER than the s-UE at an effective modulation order of 16-QAM. The proposed framework thus offers a low-complexity, SIC-free, and downlink-friendly solution for grant-free uplink in ISAC systems.

Detecting unauthorized UAV flights is critical for surveillance, security, and airspace management. Acoustic drone detection, which relies on the distinctive propeller and motor sounds of UAVs, provides a low-cost, passive solution that requires no line of sight. A central challenge is generalization: reliably distinguishing drone signatures from ambient noise across unseen recording setups, environments, and UAV types (out-of-domain). Inspired by advances in large-scale audio pretraining, we develop a compact DNN-based detector and improve its generalization by (1) pretraining the model for broad sound-event classification before fine-tuning on diverse in-house and public drone recordings, and (2) applying on-the-fly augmentations (pitch shifting, noise mixing, microphone transfer function simulation, spectrogram augmentation) to expose the model to varied acoustic conditions. An ablation study quantifies the impact of each augmentation. For evaluation, we set target false-positive rates (FPR) aligned with real-world surveillance needs and report true-positive rates (TPR) on both in-domain data (public IDMT Berne 2022) and out-of-domain data (public AuDroK). Our results show that pretraining is the dominant factor for robust detection, yielding substantial TPR improvements over training from scratch on all benchmarks. The full augmentation chain provides additional gains on acoustically mismatched out-of-domain data, achieving the best mean TPR on the AuDroK subsets and the largest improvements on the most challenging scenarios. We further validate real-world applicability by measuring false positives on public non-drone corpora (IDMT-TRAFFIC and ESC-50), demonstrating equally low FPR on unfamiliar backgrounds. A distance-dependent analysis on IDMT Berne 2022 shows effective detection at distances up to 150 m.

This paper studies model-free optimal control design and its convergence properties for linear time-invariant systems subject to probabilistic risk or chance constraints. In particular, we study a natural policy gradient (NPG)-based actor-critic (AC) algorithm with two timescales, using a Lagrangian primal-dual framework to enforce the constraint. Furthermore, the risk is defined as the probability that a function of the one-step-ahead state exceeds a user-specified threshold. To our knowledge, this is the first work to study the analytical convergence properties for NPG-based AC in a chance-constrained linear-quadratic Gaussian (LQG) regulator setting without model knowledge. We establish the coercivity and gradient dominance properties of the Lagrangian function, which ensure linear convergence and closed-loop stability during training for the actor. On the other hand, we analyse the convergence properties of the temporal difference (TD(0)) learning for the critic, applying stochastic approximation theory. Also, we demonstrate no duality gap in the constrained optimisation problem. Additionally, we have performed numerical analysis of the convergence properties and accuracy of the proposed method, comparing it with model-based chance-constrained LQR and scenario-based MPC. Results show that our approach effectively limits risk while maintaining near-optimal performance, without requiring full model knowledge or real-time optimisation.

We introduce the notion of worst-case posterior and worst-case likelihood sensitivity. These measure, respectively, the sensitivity of the expected cost to worst-case perturbations of the posterior distribution and worst-case perturbations of the likelihood of a Bayesian model. Each defines a quantitative measure of robustness. A decision maker concerned about the sensitivity of the out-of-sample expected cost to deviations from her assumptions will want a decision for which both sensitivities are small. We derive posterior and likelihood sensitivities for uncertainty sets defined in terms of deviation measures. Posterior sensitivity vanishes when the posterior variance shrinks to zero, which occurs when parameter uncertainty is eliminated from learning. Parameter learning does not eliminate likelihood sensitivity. A distributionally robust formulation of a Bayesian optimization problem makes a near-Pareto-optimal tradeoff between performance (expected cost) and robustness (posterior and likelihood sensitivity).

Undersampled magnetic resonance imaging (MRI) reconstruction seeks to recover temporally or contrast-varying image series from incomplete multicoil k-space data while preserving state-dependent fidelity for dynamic and quantitative MRI (qMRI). Existing scan-specific implicit neural representations (INRs) often use monolithic spatiotemporal coordinate fields, explicit subspaces, motion or deformation models, calibration variables, or sequence-specific quantitative signal models. These design choices can limit flexibility in sharing spatial information while adapting image synthesis across acquisition states. Moreover, many INR-based baselines remain computationally demanding, typically requiring per-scan optimization times on the order of hundreds to thousands of seconds. We propose MoE-dqINR, a scan-specific multicoil MRI reconstruction framework that factorizes the image-domain representation into shared spatial experts and a state-conditioned routing pathway. Spatial experts encode reusable coordinate-dependent image content, whereas routing weights, conditioned on ordered acquisition states, synthesize each dynamic frame or contrast state from a common expert bank. The representation is coupled to a multicoil MRI forward model, uses the normalized state index to drive routing in both dynamic and quantitative MRI. By separating shared spatial representation from state-dependent synthesis, the framework provides an image-first architecture for dynamic and quantitative MRI while reducing scan-specific INR optimization to approximately 30 s per scan in our experiments. The proposed formulation establishes state-conditioned mixture-of-experts INR as a scan-specific multicoil MRI reconstruction prior that unifies shared spatial representation, dynamic- and qMRI-specific synthesis, and practical per-scan efficiency.

To make cross-band channel prediction practical for AI-native RAN, algorithms must generalize across diverse environments and support real-time inference. Existing approaches achieve one but not both. To bridge this gap, we introduce GUIDE, a physics-guided deep unfolding framework that embeds wireless channel physics into differentiable layers. Without retraining in unseen environments, GUIDE achieves 2.75x beamforming gain than the deep learning-based baseline FIRE with only a slight increase in inference time, and 1.39x beamforming gain than the strongest model-based baseline R2F2 while running over 1610x faster.

This paper studies the control of discrete-time linear fractional-order networks, a flexible modeling framework for systems with long-range memory such as power grids, biological networks, and neuronal circuits. In contrast to the common view that fractional exponents (time-scales) are fixed parameters, we show that they can be systematically steered, together with the network coupling matrix, by appropriately designed input sequences. We first derive algebraic conditions under which the coupling matrix and the vector of fractional exponents of a given network can be reconfigured to desired values, and we characterize how truncating the infinite-memory term impacts the resulting dynamics. Building on these results, we construct an equivalent linear representation that isolates the contribution of memory, and we introduce a fractional reachability matrix that provides explicit conditions for jointly steering both network parameters and state in a finite number of steps. To address practical implementations, we further formulate an energy-constrained steering problem that incorporates actuator bounds and finite-memory approximations as a quadratic program. The framework is illustrated on low-dimensional toy examples, on larger networks with Erdos-Renyi, Barabasi-Albert, and Watts-Strogatz topologies, and on a brain network model inferred from electrocorticography recordings of an epilepsy patient, where we showcase transitions between pre-seizure and seizure configurations.

The experimental validation of diffuse scattering models has long been limited by the inability to spatially separate specular and diffuse contributions in measured channels. This paper overcomes this limitation by combining super-resolution multipath component (MPC) extraction, which resolves individual propagation paths including the specular component, with digital-twin-assisted geometry, enabling the spatial separation of specular and diffuse contributions from bistatic measurements at 28~GHz. Using this framework, we provide the first measurement-driven validation of the Effective Roughness (ER) model with independent characterization of diffuse scattering across ten common building materials, each measured over 266 angular configurations and all polarization combinations (HH, HV, VH, VV). Furthermore, we extend the ER framework by proposing a novel angle-dependent cross-polarization discrimination (XPD) model, capturing the geometry-dependent nature of depolarization that is neglected in existing approaches. The proposed method reproduces the measured diffuse power trends, achieving RMSE values as low as 3 dB across the tested materials, and improves XPD prediction over the baseline constant-XPD model for nearly all material-polarization cases. These results establish a physically consistent and practically viable approach for high-fidelity channel modeling in mmWave systems.

This study introduces a novel Air Traffic Control (ATC) concept to support self-separation between vehicles in Urban Air Mobility (UAM) corridors. Our proposed scheme involves sharing intended arrival schedules at Constrained Waypoints (CWPs) among UAM operators. We propose two approaches to assist the arrival scheduling at CWPs by computing the minimum arrival time gap necessary for each pair of vehicles to ensure their safety throughout the flights within the corridor. The first approach considers the minimum separation distance required by the Near Mid-Air-Collision (NMAC) avoidance rules, while the second one is based on the Responsibility-Sensitive Safety (RSS) rules. We demonstrate that the NMAC-rule-based approach can effectively prevent collisions in normal circumstances, where the vehicles adhere to the speed limits of the corridor. However, this approach does not guarantee safety if vehicles exceed the speed limits. Conversely, while the RSS-rule-based approach ensures collision prevention during emergencies when vehicles exceed speed limits, it may require larger arrival time gaps under normal circumstances, which may lead to reduced traffic flow. Our results are confirmed through numerical simulations.

This paper investigates distributed beam focusing for coordinated satellite constellations with phased arrays, motivated by future non-terrestrial network (NTN) systems. A geometric and channel model is developed by incorporating satellite positions, array orientations, antenna directivity, and polarization effects. Under ideal synchronization, the achievable coherent combining gain is analyzed for different constellation geometries, showing that maximum ratio transmission (MRT) enables quadratic scaling of the received power with the number of satellites. The impact of phase errors caused by residual synchronization, timing, mobility, and localization mismatches is then investigated. Closed-form expressions for the average coherent gain are derived for uniformly distributed timing offsets, demonstrating the transition from coherent to non-coherent combining. The results show that synchronization and timing mismatches reduce the coherent combining gain, while geometry dependent effects govern the resulting spatial focusing behavior. Numerical results further show that linear and circular constellations provide different focusing characteristics and spatial separation capabilities. However, MRT-based focusing results in strong sidelobes and limited spatial division capability, motivating the need for joint analog beamforming and digital precoding optimization to improve spatial selectivity and robustness against mobility and localization errors.

In public address systems and hearing aids, the maximally achievable amplification or gain is limited by acoustic feedback. Therefore, in order to be able to apply a higher gain, feedback cancellation methods are required. In addition, it is oftentimes also desirable to dereverberate a recorded signal, that is, remove the late reverberation component of the signal, before playing it back. In this paper, it is shown that under two mild conditions, the acoustic feedback signal can be written as a reverberant version of the source signal. Therefore, it is possible to treat the joint dereverberation and acoustic feedback cancellation problem as a dereverberation-only problem, meaning that dereverberation algorithms can be applied to the joint problem. Simulations corroborate this finding

Non-terrestrial networks (NTNs) are expected to play a pivotal role in sixth-generation (6G) systems by enabling ubiquitous connectivity and massive communication. In this context, channel prediction emerges as a key technique to improve the spectrum utilization efficiency by limiting the pilot overhead. However, many proposed predictors based on artificial intelligence (AI) are characterized by high inference complexity, posing challenges to onboard implementation. In this paper, we address the challenge of designing accurate yet computationally efficient channel prediction techniques tailored to low Earth orbit (LEO) NTNs, where strict power constraints limit model complexity, to enable spectral efficiency gains. We propose an iterative joint channel estimation and prediction framework in the context of 6G NTNs that significantly reduces pilot overhead by transmitting pilots only in the initial slot and relying on data-driven processing for subsequent slots. We introduce Data-driven Refinement and Iterative Forecast for wireless channel Tracking (DRIFT), a lightweight architecture that refines data-aided channel estimates and predicts future channel frequency responses with low computational cost and reduced error propagation. Two predictor variants based on convolutional and long short-term memory layers are investigated. Simulation results in an end-to-end simulation of an uplink LEO NTN scenario show that the proposed approach achieves up to 12% spectral efficiency gain compared to conventional pilot-based systems, with robustness to training-test mismatches and consistent performance across different channel models. Moreover, DRIFT requires fewer than 200k multiply-accumulate operations, making it suitable for on-board satellite implementation under stringent power constraints.

This paper characterizes the performance limits of optimal array designs using orthogonal and coherent waveforms for both linear and planar arrays. For orthogonal waveforms, we show that the single-target Cramér-Rao Bound (CRB) depends on the sum of the so-called spatial variances of the transmit (Tx) and receive (Rx) arrays, or equivalently, the spatial variance of the sum co-array weighted by the multiplicities of the virtual sensors. This reveals that CRB-optimal geometries are inherently redundant, highlighting a fundamental trade-off between mean squared error (MSE) and identifiability in parameter estimation. Moreover, we derive optimal Tx-Rx sensor allocations given a total sensor budget and show that unequal allocation (favoring the Rx) is optimal even for nonredundant arrays, questioning conventional designs. We extend our results to planar arrays, providing a new general condition that the spatial covariances of the Tx and Rx arrays should satisfy for the optimal waveforms to direct power in the target direction. Additionally, we establish a connection between Diophantine equations and array geometries with equal CRB, along with a constructive method for designing such arrays. Our work provides new guidelines for and insights into optimal array and waveform design with relevance in emerging active sensing multiple-input multiple-output systems.

Purpose: T1-weighted MPRAGE remains a cornerstone of clinical anatomical imaging, yet its long acquisition times constrain routine use. Established acceleration techniques, namely Parallel Imaging (PI) and Compressed Sensing (CS), tend to introduce substantial noise and blurring when pushed to high acceleration factors. Although they rely on fundamentally different redundancies, combining them synergistically remains an open challenge. Methods: The GoLF-SPARKLING framework was extended to jointly exploit two acceleration mechanisms: GRAPPA-based PI in the central k-space region and variable-density CS in the periphery, with independent acceleration factors in each zone. To preserve smooth signal evolution throughout the inversion-recovery period and avoid modulation artifacts, the acquisition trajectory was reordered accordingly. The resulting method was evaluated prospectively in vivo at 1mm isotropic resolution and benchmarked against Wave-CAIPI and Poisson-disk sampling. Results: The proposed hybrid approach produced sharper, less noisy, and more stable whole-brain images in approximately one minute than either acceleration strategy alone. Purely PI-based reconstructions were degraded by high g-factor noise, while purely CS-based reconstructions exhibited pronounced blurring. Furthermore, this method yielded lower average volumetric errors in downstream automated brain segmentation than state-of-the-art acceleration techniques, demonstrating its clinical utility. Conclusion: By jointly leveraging PI and CS, GoLF-SPARKLING achieves high acceleration factors that enable sub-minute, high-quality anatomical MRI. This translates into greater clinical throughput and more reliable imaging in patients who are challenging to scan.

Zero-shot text-to-speech (TTS) has improved substantially for single-speaker synthesis, yet expressive long-form multi-speaker dialogue remains difficult. A common workaround is to synthesize each turn with a monologue TTS model and stitch the outputs together. This adds inference cost and often breaks acoustic consistency, conversational coherence, and affective continuity across turns. Recent dialogue TTS systems have begun to address this setting, but they still struggle to keep expressive coherence, controllable speaker switching, and monologue quality at the same time. We present SwanData-Speech and SwanVoice. SwanData-Speech builds monologue and dialogue corpora from in-the-wild audio, using Swan Forced Aligner for pause-aware word-level alignment and RobustMegaTTS3 for pronunciation-hard cases. Built on these data, SwanVoice is a zero-shot TTS model for 1--4 speakers, combining a 25 Hz VAE, raw-text conditioning with pause-aware symbols and pinyin substitution, and a flow-matching DiT with speaker-turn conditioning. Training starts from monologue speech, moves through mixed and real dialogue data, and then uses DiffusionNFT post-training with phone-level and speaker-similarity rewards. On SwanBench-Speech, SwanVoice obtains higher richness and hierarchy scores than all evaluated open-source baselines in both monologue and dialogue settings, while content accuracy remains the main limitation. Audio demos are available at https://swanaigc.github.io//#swanvoice.

Semantic communication has demonstrated significant potential for image transmission, especially in bandwidth-limited and low signal-to-noise ratio scenarios. However, most existing methods are based on analog transmission, which poses challenges to the compatibility with existing digital communication systems. Existing digital semantic communication methods commonly adopt conventional quadrature amplitude modulation constellations, which mismatch the empirical distribution of semantic features produced by the semantic encoder. This paper proposes a distribution-aware learnable modulation for semantic communication framework, which bridges semantic feature representations and discrete modulation through constellation learning. Specifically, a learnable constellation module, initialized with an amplitude phase shift keying geometric prior, is developed to refine the constellation geometry as a trainable codebook, enabling modulation symbols to better align with the distribution of semantic features. To enable end-to-end optimization, a two-stage training strategy is introduced, combining differentiable soft assignment with straight-through estimator. Simulation results show that the proposed framework consistently outperforms existing digital semantic communication schemes and achieves performance comparable to advanced analog methods.

HyperSpectral-MultiSpectral Image (HSI-MSI) fusion enables high-resolution hyperspectral imaging by combining the rich spectral information of low-spatial-resolution hyperspectral images with the detailed spatial structure of multispectral images. Classical methods such as Coupled Nonnegative Matrix Factorization (CNMF) benefit from a strong physical interpretability but suffer from inferior results compared to their deep-learning counterparts. To address this limitation, we propose SCALMU (Synthetically-trained Coupling of Adaptive Learned Multiplicative Updates), a novel unrolled neural network architecture that integrates adaptive learnable matrices within the classical framework of CNMF multiplicative updates, improving its results. Due to its architectural proximity with CNMF, the resulting algorithm preserves physical interpretability and nonnegativity constraints. To overcome data scarcity for training, we additionally generate a synthetic HSI-MSI dataset via the dead leaves model, enabling synthetic supervision. SCALMU is then trained end-to-end on this dataset. Experiments demonstrate SCALMU's superiority over state-of-the-art methods on several datasets. The code is available at https://github.com/xinxinxu99/SCALMU.git

Recent advancements in text-guided audio generation have yielded promising results in diverse domains, including sound effects, speech, and music. However, jointly generating speech with environmental audio remains challenging due to the inherent disparities in their acoustic patterns and temporal dynamics. We propose ImmersiveTTS, an environment-aware text-to-speech (TTS) model that generates natural speech seamlessly integrated within environmental contexts by explicitly modeling cross-modal interactions. Our model builds on a multimodal diffusion transformer and fuses transcript-aligned speech latent with text-conditioned environmental context via joint attention. To enhance semantic consistency, we introduce a domain-specific representation alignment objective tailored to environment-aware TTS, leveraging complementary self-supervised representations from speech and audio encoders. Experimental results show that ImmersiveTTS achieves higher naturalness, intelligibility, and audio fidelity than existing approaches across objective metrics and human listening tests.

Real-time and accurate spatial audio generation is pivotal for delivering an immersive experience. However, existing spatial audio synthesis technologies are often encumbered by a tradeoff between generation quality and high inference latency, as well as difficulty in capturing precise spatial information from multimodal inputs. To address these challenges, we propose SwanSphere, a unified streaming framework for high-fidelity spatial audio generation from panoramic videos and text prompts. SwanSphere mainly makes the following contributions: 1) We introduce a causal autoregressive diffusion transformer architecture that enables streaming high-quality spatial audio generation. 2) We design a Spatial Video-Audio Contrastive (SVAC) learning strategy to align the video encoder with the acoustic domain, and further employ a multi-objective online direct preference optimization (ODPO) scheme, resulting in strong spatial perception and robust multimodal spatial audio synthesis. 3) To alleviate the current scarcity of spatial audio datasets, we also develop an automated annotation pipeline for generating detailed spatial captions. Experimental results demonstrate that SwanSphere achieves superior performance in both video-to-spatial and text-to-spatial audio generation tasks. Demos can be found at: https://swanaigc.github.io.