Papers

We introduce SSMRadNet, the first multi-scale State Space Model (SSM) based detector for Frequency Modulated Continuous Wave (FMCW) radar that sequentially processes raw ADC samples through two SSMs. One SSM learns a chirp-wise feature by sequentially processing samples from all receiver channels within one chirp, and a second SSM learns a representation of a frame by sequentially processing chirp-wise features. The latent representations of a radar frame are decoded to perform segmentation and detection tasks. Comprehensive evaluations on the RADIal dataset show SSMRadNet has 10-33x fewer parameters and 60-88x less computation (GFLOPs) while being 3.7x faster than state-of-the-art transformer and convolution-based radar detectors at competitive performance for segmentation tasks.

From organisms to machines, autonomous systems rely on measured sensory cues to estimate unknown information about themselves or their environment. For nonlinear systems, strategic sensor motion can be leveraged to extract otherwise inaccessible information. This principle, known as active sensing, is widespread in biology yet difficult to study, and remains underutilized in engineered systems due to the challenge of systematically designing active sensing motifs. Here, we introduce the method “BOUNDS: Bounding Observability for Uncertain Nonlinear Dynamic Systems", and Python package pybounds, which can discover movement motifs that increase the information encoded in sensory cues. To exploit sporadic estimates from bouts of active sensing, we further introduce the Augmented Information Kalman Filter (AI-KF). The AI-KF uses insight from BOUNDS to dynamically fuse neural network and model-based estimation. We demonstrate BOUNDS and the AI-KF on a flying agent model and experimental GPS-denied data from a quadcopter, revealing how specific active movements improve estimates of ground speed, altitude, and wind direction. Altogether, our work will prove useful for designing sensor-minimal autonomous systems and investigating active sensing in living organisms.

Data centers are facilities housing computing infrastructure for processing and storing digital information. The rapid expansion of artificial intelligence is driving unprecedented growth in data center capacity, with global electricity demand from data centers projected to double by 2026. This growth creates substantial challenges for power transmission networks, as large concentrated loads can cause congestion and threaten grid reliability. Meanwhile, the intermittent nature of solar and wind generation requires flexible resources to maintain grid reliability and minimize curtailment. This paper assesses whether data center spatial flexibility-the ability to migrate computational workloads geographically-can serve as a grid resource to address these challenges. An optimal power flow model is developed to co-optimize generation dispatch, security reserves, and flexible data center loads. Case studies on a modified IEEE 73-bus system show that inflexible data center placement can lead to severe transmission violations, with line overloads reaching 30.1%. Enabling spatial flexibility mitigates these violations in the studied scenarios and restores system feasibility. This flexibility also reduces solar curtailment by up to 61.0% by strategically reallocating load to solar-rich areas. The results suggest that spatial flexibility offers a viable approach to defer transmission upgrades and enhance renewable utilization.

This work presents a global-to-local, task-aware fault detection and identification (FDI) framework for multi-spacecraft systems conducting collaborative inspection missions in low Earth orbit. The inspection task is represented by a global information-driven cost functional that integrates the sensor model, spacecraft poses, and mission-level information-gain objectives. This formulation links guidance, control, and FDI by using the same cost function to drive both global task allocation and local sensing or motion decisions. Fault detection is achieved through comparisons between expected and observed task metrics, while higher-order cost-gradient measures enable the identification of faults among sensors, actuators, and state estimators. An adaptive thresholding mechanism captures the time-varying inspection geometry and dynamic mission conditions. Simulation results for representative multi-spacecraft inspection scenarios demonstrate the reliability of fault localization and classification under uncertainty, providing a unified, information-driven foundation for resilient autonomous inspection architectures.

The alternating direction method of multipliers (ADMM) is a powerful algorithm for solving decentralized optimization problems including networked microgrid energy management (NetMEM). However, its performance is highly sensitive to the selection of its penalty parameter h̊o}, which can lead to slow convergence, suboptimal solutions, or even algorithm divergence. This paper evaluates and compares three district ADMM formulations to solve the NetMEM problem, which explore different methods to determine appropriate stopping points, aiming to yield high-quality solutions. Furthermore, an adaptive penalty heuristic is also incorporated into each method to analyze its potential impact on ADMM performance. Different case studies on networks of varying sizes demonstrate that an objective-based ADMM approach, denominated as OB-ADMM, is significantly more robust to the choice of h̊o}, consistently yielding solutions closer to the centralized optimal benchmark by preventing premature algorithm stopping.

Ensuring the safety of real-world systems is challenging, especially when they rely on learned perception modules to infer the system state from high-dimensional sensor data. These perception modules are vulnerable to epistemic uncertainty, often failing when encountering out-of-distribution (OoD) measurements not seen during training. To address this gap, we introduce ATOM-CBF (Adaptive-To-OoD-Measurement Control Barrier Function), a novel safe control framework that explicitly computes and adapts to the epistemic uncertainty from OoD measurements, without the need for ground-truth labels or information on distribution shifts. Our approach features two key components: (1) an OoD-aware adaptive perception error margin and (2) a safety filter that integrates this adaptive error margin, enabling the filter to adjust its conservatism in real-time. We provide empirical validation in simulations, demonstrating that ATOM-CBF maintains safety for an F1Tenth vehicle with LiDAR scans and a quadruped robot with RGB images.

Mobility systems are complex socio-technical environments influenced by multiple stakeholders with hierarchically interdependent decisions, rendering effective control and policy design inherently challenging. We bridge hierarchical game-theoretic modeling with online feedback optimization by casting urban mobility as a tri-level Stackelberg game (travelers, operators, municipality) closed in a feedback loop. The municipality iteratively updates taxes, subsidies, and operational constraints using a projected two-point (gradient-free) scheme, while lower levels respond through equilibrium computations (Frank-Wolfe for traveler equilibrium; operator best responses). This model-free pipeline enforces constraints, accommodates heterogeneous users and modes, and scales to higher-dimensional policy vectors without differentiating through equilibrium maps. On a real multimodal network for Zurich, Switzerland, our method attains substantially better municipal objectives than Bayesian optimization and Genetic algorithms, and identifies integration incentives that increase multimodal usage while improving both operator objectives. The results show that feedback-based regulation can steer competition toward cooperative outcomes and deliver tangible welfare gains in complex, data-rich mobility ecosystems.

Speech-to-Speech (S2S) models have shown promising dialogue capabilities, but their ability to handle paralinguistic cues - such as emotion, tone, and speaker attributes - and to respond appropriately in both content and style remains under-explored. Progress is further hindered by the scarcity of high-quality and expressive demonstrations. To address this, we introduce a new reinforcement learning (RL) framework for paralinguistic-aware S2S, ParaS2S, which evaluates and optimizes both response content and speaking style directly at the waveform level. We first construct ParaS2SBench, a benchmark that evaluates the naturalness of input-output pairs in terms of content and speaking style using expressive and challenging queries. For the automatic judge, we propose a PolyTone training strategy and a multi-stage framework, preventing the style hallucination of end-to-end audio LLM judging. Our judge correlates well with human preferences and is scalable, enabling the model to interact and learn from unlabeled speech via RL. Experiments show that existing S2S models fail to respond appropriately to paralinguistic attributes, performing no better than pipeline-based baselines. Our RL approach (ParaS2SAlign) achieves a 10% relative improvement in the appropriateness of response content and speaking style on ParaS2SBench over supervised fine-tuning (SFT), surpassing all prior models while requiring substantially fewer paired demonstrations than pure SFT. Our findings highlight the need for a scalable and accurate automatic evaluator for speech-to-speech interaction.

We study the energy efficiency of pinching-antenna systems (PASSs) by developing a consistent formulation for power distribution in these systems. The per-antenna power distribution in PASSs is not controlled explicitly by a power allocation policy, but rather implicitly through tuning of pinching couplings and locations. Both these factors are tunable: (i) pinching locations are tuned using movable elements, and (ii) couplings can be tuned by varying the effective coupling length of the pinching elements. While the former is feasible to be addressed dynamically in settings with low user mobility, the latter cannot be addressed at a high rate. We thus develop a class of hybrid dynamic-static algorithms, which maximize the energy efficiency by updating the system parameters at different rates. Our experimental results depict that dynamic tuning of pinching locations can significantly boost energy efficiency of PASSs.

In this letter, we formulate a compositional distributed learning framework for multi-view perception by leveraging the maximal coding rate reduction principle combined with subspace basis fusion. In the proposed algorithm, each agent conducts a periodic singular value decomposition on its learned subspaces and exchanges truncated basis matrices, based on which the fused subspaces are obtained. By introducing a projection matrix and minimizing the distance between the outputs and its projection, the learned representations are enforced towards the fused subspaces. It is proved that the trace on the coding-rate change is bounded and the consistency of basis fusion is guaranteed theoretically. Numerical simulations validate that the proposed algorithm achieves high classification accuracy while maintaining representations' diversity, compared to baselines showing correlated subspaces and coupled representations.

Now that disease-modifying therapies for Alzheimer disease have been approved by regulatory agencies, the early, objective, and accurate clinical diagnosis of AD based on the lowest-cost measurement modalities possible has become an increasingly urgent need. In this study, we propose a novel feature extraction method using persistent homology to analyze structural MRI of the brain. This approach converts topological features into powerful feature vectors through Betti functions. By integrating these feature vectors with a simple machine learning model like XGBoost, we achieve a computationally efficient machine learning model. Our model outperforms state-of-the-art deep learning models in both binary and three-class classification tasks for ADNI 3D MRI disease diagnosis. Using 10-fold cross-validation, our model achieved an average accuracy of 97.43 percent and sensitivity of 99.09 percent for binary classification. For three-class classification, it achieved an average accuracy of 95.47 percent and sensitivity of 94.98 percent. Unlike many deep learning models, our approach does not require data augmentation or extensive preprocessing, making it particularly suitable for smaller datasets. Topological features differ significantly from those commonly extracted using convolutional filters and other deep learning machinery. Because it provides an entirely different type of information from machine learning models, it has the potential to combine topological features with other models later on.

Physics-Informed Neural Networks (PINNs) have emerged as a promising framework for solving forward and inverse problems governed by differential equations. However, their reliability when used in ill-posed inverse problems remains poorly understood. In this study, we explore the fundamental limitations of PINNs using a simple illustrative case: RC low-pass filters. Showing that while PINNs can accurately predict system dynamics in forward problems, they fail to recover unique physical parameters when solving inverse problems when more than two parameters are approximated. Our findings provide grounds to understand the boundaries of PINNs applicability for parameter discovery in physical systems.

Microbial consortia offer significant biotechnological advantages over monocultures for bioproduction. However, industrial deployment is hampered by the lack of scalable architectures to ensure stable coexistence between populations. Existing strategies rely on genetic modifications, which impose metabolic load, or environmental changes, which can reduce production. We present a versatile control architecture to regulate density and composition of a two-strain consortium without genetic engineering or drastic environmental changes. Our bioreactor-based control architecture comprises a mixing chamber where both strains are co-cultured and a reservoir sustaining the slower-growing strain. For both chambers we develop model-based and sim-to-real learning controllers. The control architecture is then validated in vivo on a two-strain Escherichia coli consortium, achieving precise and robust regulation of consortium density and composition, including tracking of time-varying references and recovery from perturbations.

Nyquist Signaling Modulations (NSMs) are a new signaling paradigm inspired by faster-than-Nyquist principles but based on a distinct approach that enables controlled inter-symbol interference through carefully designed finite-impulse-response filters. NSMs can operate in any number of dimensions, including mixed-dimensional configurations, offering wide flexibility in filter design, optional energy balancing, and preservation of the 2-ASK minimum squared Euclidean distance (MSED). Both real and rational tapped filters are investigated, and closed-form expressions are derived for the optimal real-tap filters in the one-dimensional case (MS-PRS), providing analytical insight and strong agreement with simulated bit-error behavior across wide SNR ranges. The paradigm is conceptually expanded through an analog Low-Density Generator Matrix (LDGM) formulation, which broadens the NSM family and unifies modulation and coding within a single, structurally coherent framework. When combined with LDPC coding, it enables efficient and naturally synergistic interaction between the analog modulation and the digital LDPC code. Alternatively, when analog LDGM is employed for both source coding and modulation, a simple and harmonious joint source-channel-modulation structure emerges. In both configurations, the constituent blocks exhibit dual graph-based architectures suited to message passing, achieving high flexibility and complexity-efficient operation. Collectively, these results establish promising physical-layer directions for future 6G communication systems.

This paper presents an asymptotic analysis of Multiple-Input Multiple-Output (MIMO) systems assisted by a 1-bit Reconfigurable Intelligent Surface (RIS) under Ricean fading conditions. Using random matrix theory, we show that, in the asymptotic regime, the dominant singular values and vectors of the transmitter-RIS and RIS-receiver channels converge to their deterministic Line-of-Sight (LoS) components, almost irrespective of the Ricean factors. This enables RIS phase configuration using only LoS information through a closed-form Sign Alignment (SA) rule that maximizes the channel gain. Furthermore, when the RIS is asymptotically larger than the transceiver arrays, proper RIS configuration can render the end-to-end MIMO channel in the capacity formula asymptotically diagonal, thereby eliminating inter-stream interference and enabling Over-The-Air (OTA) spatial multiplexing without channel knowledge at the transmitter. Building on this result, a waterfilling-inspired SA algorithm that allocates RIS elements to spatial streams, based on the asymptotic singular values and statistical channel parameters, is proposed. Simulation results validate the theoretical analyses, demonstrating that the proposed schemes achieve performance comparable to conventional Riemannian manifold optimization, but with orders of magnitude lower runtime.

Most works on pilot-aided orthogonal time frequency space (OTFS) channel estimation operate in the delay-Doppler (DD) domain, where embedded-pilot guard regions must cover the unknown maximum delay-Doppler spread, causing high overhead that worsens with antenna scaling: per-antenna guard regions increase overhead with the number of transmit antennas, while shared guard regions avoid this but introduce pilot contamination. Fractional Doppler further reduces DD-domain sparsity and increases estimation complexity. This paper shows that the time-frequency (TF) representation provides a low-overhead, scalable framework for OTFS channel estimation based on physical scatterer parameters (angle, delay, Doppler, and gain). We derive exact TF- and DD-domain input-output relationships under fractional Doppler. The TF expression decomposes the received signal into a desired component, identical in form for integer and fractional Doppler, and an explicitly characterized interference term; this Doppler-invariant structure enables scalable TF-domain estimation. This motivates a TF-domain pilot design using private TF bins protected by TF/DD guard bins, preserving cross-antenna pilot orthogonality with overhead that depends only on the number of pilots, not the array size. We then develop a low-complexity coarse-to-fine estimation method combining discrete Fourier transform (DFT)-based coarse estimation with dimensionality-reduced sparse recovery. The exact DD-domain expression provides the signal model for data-symbol recovery and explains why DD-domain estimation complexity grows substantially under fractional Doppler. Simulations show accurate channel state information (CSI) estimation under integer and fractional Doppler, robustness to pilot pollution, and scalability to large arrays with substantially lower overhead and complexity than existing schemes.

Autonomous systems increasingly rely on human feedback to align their behavior, expressed as pairwise comparisons, rankings, or demonstrations. While existing methods can adapt behaviors, they often fail to guarantee safety in safety-critical domains. We propose a safety-guaranteed, optimal, and efficient approach for solving the learning problem from preferences, rankings, or demonstrations using Weighted Signal Temporal Logic (WSTL). WSTL learning problems, when implemented naively, lead to multi-linear constraints in the weights to be learned. By introducing structural pruning and log-transform procedures, we reduce the problem size and recast it as a Mixed-Integer Linear Program while preserving safety guarantees. Experiments on robotic navigation and real-world Formula 1 data demonstrate that the method captures nuanced preferences and models complex task objectives.

Despite the growing number of automated vehicles on public roads, operating such systems in open contexts inevitably involves incidents. Developing a defensible case that the residual risk is reduced to a reasonable (societally acceptable) level is hence a prerequisite to be prepared for potential liability cases. A "safety argumentation" is a common means to represent this case. In this paper, we contribute to the state of the art in terms of process guidance on argumentation creation and maintenance - aiming to promote a safety-argumentation-by-design paradigm, which mandates co-developing both the system and argumentation from the earliest stages. Initially, we extend a systematic design model for automated driving functions with an argumentation layer to address prevailing misconceptions regarding the development of safety arguments in a process context. Identified limitations of this extension motivate our complementary design of a dedicated argumentation life cycle that serves as an additional process viewpoint. Correspondingly, we define literature- and expert-based process requirements. To illustrate the safety argumentation life cycle that we propose as a result of implementing these consolidated requirements, we demonstrate principles of the introduced process phases (baselining, evolution, continuous maintenance) by an argumentation example on an operational design domain exit response.

Zero-shot singing voice conversion (SVC) transforms a source singer's timbre to an unseen target speaker's voice while preserving melodic content without fine-tuning. Existing methods model speaker timbre and vocal content separately, losing essential acoustic information that degrades output quality while requiring significant computational resources. To overcome these limitations, we propose HQ-SVC, an efficient framework for high-quality zero-shot SVC. HQ-SVC first extracts jointly content and speaker features using a decoupled codec. It then enhances fidelity through pitch and volume modeling, preserving critical acoustic information typically lost in separate modeling approaches, and progressively refines outputs via differentiable signal processing and diffusion techniques. Evaluations confirm HQ-SVC significantly outperforms state-of-the-art zero-shot SVC methods in conversion quality and efficiency. Beyond voice conversion, HQ-SVC achieves superior voice naturalness compared to specialized audio super-resolution methods while natively supporting voice super-resolution tasks.

According to the recent 3GPP decisions on 6G air interface, orthogonal frequency-division multiplexing (OFDM)-based waveforms are the primary candidates for future integrated sensing and communication (ISAC) systems. In this paper, we consider a monostatic sensing scenario in which OFDM is used for the downlink and its reflected echo signal is used for sensing. OFDM and discrete Fourier transform-spread OFDM (DFT-s-OFDM) are the options for uplink transmission. When OFDM is used in the uplink, the power difference between this signal and the echo signal leads to a power-domain non-orthogonal multiple access (PD-NOMA) scenario. In contrast, adopting DFT-s-OFDM as uplink signal enables a waveform-domain NOMA(WD-NOMA). Affine frequency-division multiplexing (AFDM) and orthogonal time frequency space (OTFS) have been proven to be DFT-s-OFDM based waveforms. This work focuses on such a WD-NOMA system, where AFDM or OTFS is used as uplink waveform and OFDM is employed for downlink transmission and sensing. We show that the OFDM signal exhibits additive white Gaussian noise (AWGN)-like behavior in the affine domain, allowing it to be modeled as white noise in uplink symbol detection. To enable accurate data detection performance, an AFDM frame design and a noise power estimation (NPE) method are developed. Furthermore, a two-dimensional orthogonal matching pursuit (2D-OMP) algorithm is applied for sensing by iteratively identifying delay-Doppler components of each target. Simulation results demonstrate that the WD-NOMA ISAC system, employing either AFDM or OTFS, outperforms the PD-NOMA ISAC system that uses only the OFDM waveform in terms of bit error rate (BER) performance. Furthermore, the proposed NPE method yields additional improvements in BER.

Encoding static images into spike trains is a fundamental step for enabling Spiking Neural Networks (SNNs) to process visual information. However, widely used methods such as rate coding, Poisson encoding, and time-to-first-spike (TTFS) often neglect spatial correlations and produce temporally inconsistent spike patterns, limiting both efficiency and interpretability. In this work, we propose a novel cluster-based encoding framework that explicitly preserves semantic structure across both spatial and temporal domains. The method first introduces a 2D spatial clustering mechanism, which leverages connected component analysis and local density estimation to identify salient foreground regions. Building upon this, we extend the approach to a 3D spatio-temporal (ST3D) encoding scheme that incorporates temporal neighborhood information, generating spike trains with enhanced temporal coherence. Experiments on the N-MNIST dataset demonstrate that the proposed ST3D encoder achieves 98.17% classification accuracy using a simple single-layer SNN, outperforming conventional TTFS encoding (97.58%). Notably, this performance is achieved with significantly fewer spikes (3800 vs. 5000 per sample), highlighting improved efficiency without sacrificing accuracy. These results indicate that the proposed method provides an interpretable, structure-aware, and computationally efficient encoding strategy, offering strong potential for neuromorphic computing applications.

Proximal methods such as the Alternating Direction Method of Multipliers (ADMM) are effective at solving constrained quadratic programs (QPs). To tackle infeasible QPs, slack variables are often introduced to ensure feasibility, which changes the structure of the problem, increases its size, and slows down numerical resolution. In this letter, we propose a simple ADMM scheme to tackle QPs with slack variables without increasing the size of the original problem. The only modification is a slightly different projection in the z-update, while the rest of the algorithm remains standard. We prove that the method is equivalent to applying ADMM to the QP with additional slack variables, even though slack variables are not added. Numerical experiments show speedups of the approach.

How complex collective behavior emerges from individual interactions is a fundamental scientific question, but experimental cost and difficulty of simultaneous multi-brain recordings limit direct study in animals. Here we introduce a novel computational framework modeling weakly electric fish-like agents with biophysically inspired electrosensing and actuation, trained to forage collectively via multi-agent reinforcement learning (MARL). Trained agents reproduce hallmarks of real fish, including curvilinear homing trajectories and heavy-tailed electric organ discharge (EOD) interval statistics, while exhibiting emergent active sensing, social foraging, dominance-like asymmetries, and aggression. We perform in silico interventions including sensor ablations, EOD silencing, and food distribution changes to identify causal drivers of social foraging. Analyses of recurrent neural dynamics further show robust encoding of task-relevant variables and social context. Our work has broad implications for the neuroethology of weakly electric fish and other social animals where extensive multi-individual neural recordings, and thus traditional data-driven modeling, remain challenging.

Diffusion and flow-matching have emerged as powerful methodologies for generative modeling, with remarkable success in capturing complex data distributions and enabling flexible guidance at inference time. Many downstream applications, however, demand enforcing hard constraints on generated samples (for example, robot trajectories must avoid obstacles), a requirement that goes beyond simple guidance. Prevailing projection-based approaches constrain the entire sampling path to the constraint manifold, which is overly restrictive and degrades sample quality. In this paper, we introduce a novel framework that reformulates hard-constrained sampling as a trajectory optimization problem. Our key insight is to leverage numerical optimal control to steer the sampling trajectory so that constraints are satisfied precisely at the terminal time. By exploiting the underlying structure of flow-matching models and adopting techniques from model predictive control, we transform this otherwise complex constrained optimization problem into a tractable surrogate that can be solved efficiently and effectively. Furthermore, this trajectory optimization perspective offers significant flexibility beyond mere constraint satisfaction, allowing for the inclusion of integral costs to minimize distribution shift and terminal objectives to further enhance sample quality, all within a unified framework. We provide a control-theoretic analysis of our method, establishing bounds on the approximation error between our tractable surrogate and the ideal formulation. Extensive experiments across diverse domains, including robotics (planning), partial differential equations (boundary control), and vision (text-guided image editing), demonstrate that our algorithm, which we name $\textit{HardFlow}$, substantially outperforms existing methods in both constraint satisfaction and sample quality.

The Scaled Relative Graph (SRG) is a promising tool for stability and robustness analysis of multi-input multi-output systems. In this paper, we provide tools for exact and computable constructions of the SRG for closed linear operators, based on maximum and minimum gain computations. The results are suitable for bounded and unbounded operators, and we specify how they can be used to draw SRGs for the typical operators that are used to model linear-time-invariant dynamical systems. For the special case of state-space models, we show how the Bounded Real Lemma can be used to construct the SRG. Furthermore, we give an example of how stability analysis of dynamical systems can be done with SRGs.

Safety in stochastic control systems, which are subject to random noise with a known probability distribution, aims to compute policies that satisfy predefined operational constraints with high confidence throughout the uncertain evolution of the state variables. The unpredictable evolution of state variables poses a significant challenge for meeting predefined constraints using various control methods. To address this, we present a new algorithm that computes safe policies to determine the safety level across a finite state set. This algorithm reduces the safety objective to the standard average reward Markov Decision Process (MDP) objective. This reduction enables us to use standard techniques, such as linear programs, to compute and analyze safe policies. We validate the proposed method numerically on the Double Integrator and the Inverted Pendulum systems. Results indicate that the average-reward MDPs solution is more comprehensive, converges faster, and offers higher quality compared to the minimum discounted-reward solution.

Semantic communications mark a paradigm shift from bit-accurate transmission toward meaning-centric communication, essential as wireless systems approach theoretical capacity limits. The emergence of generative AI has catalyzed generative semantic communications, where receivers reconstruct content from minimal semantic cues by leveraging learned priors. Among generative approaches, diffusion models stand out for their superior generation quality, stable training dynamics, and rigorous theoretical foundations. However, the field currently lacks systematic guidance connecting diffusion techniques to communication system design, forcing researchers to navigate disparate literatures. This article provides the first comprehensive tutorial on diffusion models for generative semantic communications. We present score-based diffusion foundations and systematically review three technical pillars: conditional diffusion for controllable generation, efficient diffusion for accelerated inference, and generalized diffusion for cross-domain adaptation. In addition, we introduce an inverse problem perspective that reformulates semantic decoding as posterior inference, bridging semantic communications with computational imaging. Through analysis of human-centric, machine-centric, and agent-centric scenarios, we illustrate how diffusion models enable extreme compression while maintaining semantic fidelity and robustness. By bridging generative AI innovations with communication system design, this article aims to establish diffusion models as foundational components of next-generation wireless networks and beyond.

The multi-phase inverter has become more complicated, particularly in an Electric Vehicle (EV)'s power train, which requires a robust fault protection system. The proposed active short circuit and safe discharge mechanisms are also included in this work, dedicated to multi-phase converters in failure conditions. With silicon carbide (SiC) power modules increasingly used in high efficiency and high-power applications, the reliability under fault conditions is an extremely important factor. Cascading failures and permanent damage will occur in multi phase inverter systems if short circuit faults are not prevented. The proposed method combines one centralized short circuit detection, active phase shorting and controlled discharge to make these structures more robust. The on chip active short circuit mechanism isolates the affected phases quickly preventing faults from spreading to other areas of the inverter and the safe discharge mechanism controls energy discharged in fault scenarios, which reduces the thermal stress placed on essential components. The experimental results show that the proposed mechanisms can effectively enhance a fault detection performance, system response during faults, and the operation as whole at faults over the several existing methods. These mechanisms are demonstrated to be very important for enhancing the safety and reliability of multiphase inverters, especially for critical applications of such inverters as EV where high operational security is required.

Speech Foundation Models have gained significant attention recently. Prior works have shown that the fusion of representations from multiple layers of the same model or the fusion of multiple models can improve performance on downstream tasks. We unify these two fusion strategies by proposing an interface module that enables fusion across multiple upstream speech models while integrating information across their layers. We conduct extensive experiments on different self-supervised and supervised models across various speech tasks, including ASR and paralinguistic analysis, and demonstrate that our method outperforms prior fusion approaches. We further analyze its scalability concerning model size and count, highlighting the importance of selecting appropriate upstream models. Our results show that the proposed interface provides an additional performance boost when given a suitable upstream model selection, making it a promising approach for utilizing Speech Foundation Models.

Radar-based indoor 3D human pose estimation typically relied on fine-grained 3D keypoint labels, which are costly to obtain especially in complex indoor settings involving clutter, occlusions, or multiple people. In this paper, we propose RAPTR (RAdar Pose esTimation using tRansformer) under weak supervision, using only 3D BBox and 2D keypoint labels which are considerably easier and more scalable to collect. Our RAPTR is characterized by a two-stage pose decoder architecture with a pseudo-3D deformable attention to enhance (pose/joint) queries with multi-view radar features: a pose decoder estimates initial 3D poses with a 3D template loss designed to utilize the 3D BBox labels and mitigate depth ambiguities; and a joint decoder refines the initial poses with 2D keypoint labels and a 3D gravity loss. Evaluated on two indoor radar datasets, RAPTR outperforms existing methods, reducing joint position error by $34.3\%$ on HIBER and $76.9\%$ on MMVR. Our implementation is available at https://github.com/merlresearch/radar-pose-transformer.