Papers

In mixed traffic, decision-making for autonomous vehicles (AVs) confronts three interrelated challenges. First, physics-based priors incorporated into reinforcement learning (RL) models fail to capture latent interactive vehicle intentions and diverse driver behaviors, limiting the proactive reasoning capabilities. Second, abrupt maneuvers by surrounding vehicles cause non-stationarity, leaving long-tail safety events under-explored. Third, hybrid action spaces destabilize unified RL training due to the different temporal scales of continuous car-following and discrete lane-changing maneuvers. To address these issues, we propose Knowledge-Data Dual-driven Reinforcement Learning (KDDRL). First, a conditional deep generative model synthesizes intention-aware future trajectories, converting passive perception into proactive predictive states. Second, a knowledge-data dual-driven paradigm operates on these predictive states, fusing probabilistic data-driven insights with physical constraints to guide safe exploration through safety-critical scenarios. Third, a coupling module compresses both intention-aware trajectories and physical constraints into compact shared embeddings. This unified representation enables asynchronous multi-timescale optimization of continuous car-following and discrete lane-changing while preserving mutual information. Evaluations on dataset-calibrated simulations demonstrate that KDDRL effectively handles intention uncertainty, accelerates training convergence, and outperforms conventional baseline methods in terms of safety, efficiency, and comfort.

Timely urban-canopy information is essential for linking remote sensing with heat, mobility, and neighborhood planning. We developed an optical GeoAI workflow for Davis, California, using 2022 National Agriculture Imagery Program imagery (0.6 m RGB+NIR). DeepForest generated crown candidates; an NDVI threshold, non-maximum suppression, and box-prompted Segment Anything Model (ViT-B) produced a crown-anchored canopy surface. Analyses used the 25.92 km2 Census TIGER municipal boundary and a 100 m grid. The workflow retained 11,741 candidate crowns and mapped 7.71 km2 of canopy (29.8% of the city). On the identical extent, pixel precision was 0.804, pixel recall was 0.873, and 97.4% of candidate centers agreed with the 2022 USDA/CAL FIRE LiDAR-assisted canopy product (IoU 0.719; Dice 0.837; area recovery 108.5%). Approximately 49% of candidates occurred within 15 m of a road. Canopy was inversely associated with Landsat land-surface temperature (Spearman rho = -0.477; partial rho = -0.551 controlling for built probability), and spatial-lag modeling confirmed clear neighborhood structure. Two transparent attention surfaces combined canopy need with thermal and contextual indicators. The framework provides a reproducible, updateable screening layer that complements structural canopy products and municipal inventories while retaining assumptions, data provenance, and spatial diagnostics for planning interpretation.

PLEASE CHECK AND REFER TO THE PUBLISHED VERSION OF THIS PAPER ON https://doi.org/10.5334/tismir.357 This paper presents the MPB Corpus, a dataset of Brazilian Popular Music (Musica Popular Brasileira, MPB) comprising derived symbolic annotations for 500 pieces across four musical parameters – melodic contour, melodic rhythm, harmony, and the melody-harmony relationship – making it the most comprehensive dataset to date for computational musicology of Brazilian music. It is released under a CC-BY-4.0 license in an open, machine-readable tabular format, with documentation and reproducible analysis code. Because the annotations were derived from copyrighted musical scores, the corpus contains no audio recordings or score-level representations such as MIDI or MusicXML; it nonetheless provides a reusable resource for MIR research on a culturally significant repertoire that remains underrepresented in existing datasets. To support the encoding process, we introduce analytical models for capturing rhythmic and melodic information, along with tailored visualizations and summary metrics. Finally, we present a brief exploratory analysis illustrating the dataset's potential for formulating and addressing musicological questions about the genre.

This paper presents a control barrier function-based method for real-time in-domain congestion control of the Lighthill-Whitham-Richards traffic model. Traffic congestion is formulated as a distributed safety control problem, leading to a infinite-dimensional optimization problem. Through the discretization and Karush-Kuhn-Tucker (KKT) analysis, the problem is converted into a high-dimensional quadratic program. A structure-exploiting primal-dual active set algorithm is then developed to compute the safe control input in real time, with convergence guarantees. Numerical simulations with different nominal controllers demonstrate the effectiveness and real-time feasibility of the proposed approach.

Cross subject emotion decoding from electroencephalography EEG requires representations that accommodate individual variability while preserving spatial spectral structure for interpretation. This study introduces EmoDiPyraTrans, a differential graph Transformer that integrates adaptive graph recurrence, differential attention, pyramid fusion and distribution regularization over sequential relative power spectral density graphs. Across SEED, FACED, MAHNOB HCI, DEAP and DREAMER, the model achieved the highest participant mean accuracy and positive class F1 among the evaluated methods, with accuracy and F1 both reaching 0.928 on SEED. On DEP EEG, positive versus neutral accuracy reached 0.802 within healthy controls and 0.704 within participants with depression, compared with 0.591 under healthy to depression transfer and 0.581 with mixed population development. Complementary SEED analyses identified distributed spatial weighting and an alpha centred spectral preference, while configurations averaging six channels retained near full performance. These findings link generalization assessment with model derived candidate signatures to support interpretable EEG emotion decoding, with code available at https://github.com/hdy6438/EmoDiPyraTrans.

Large Audio-Language Models (LALMs) have recently shown strong capabilities in speech understanding and question answering (QA), but they also inherit privacy risks from large-scale training data, including the unintended memorization of sensitive information. In this work, we study machine unlearning for speech QA in LALMs, a setting that is more challenging than prior work on text-based Large Language Models (LLMs) or Automatic Speech Recognition (ASR) due to the tight coupling between acoustic perception and factual knowledge. We present and evaluate multiple unlearning strategies, including gradient ascent, task arithmetic, and alignment-based fine-tuning methods that enforce safe refusal responses, to remove private knowledge while still preserving performance on core capabilities. Through extensive experiments on speech QA datasets, we show that these unlearning methods can reduce the privacy leakage rate by up to 80% while maintaining near-neutral performance on non-private speech QA and general speech understanding benchmarks.

Magnetoencephalography (MEG) source localization is an ill-posed inverse problem because distinct cortical source configurations can produce similar sensor-level fields. We formulate sparse multi-source localization as a quadratic unconstrained binary optimization (QUBO) problem combined with residual-aware candidate screening. Candidate source-location groups are generated from the sensor-space residual, fixed sensor-space templates are estimated for the resulting candidates, and active templates are jointly selected using data-fit, pairwise template interactions, and soft-cardinality terms. We evaluate the method using classical simulated annealing in controlled synthetic MEG simulations, primarily under a two-source condition, and compare it with MNE, dSPM, MxNE, LCMV, and RAP-MUSIC. Across 100 main-benchmark trials, QUBO achieved a mean cardinality-aware localization error of 8.45 mm, compared with 22.35 mm for MxNE, the best-performing baseline according to this metric, corresponding to a 62.2% reduction. The composite metric adds a 50 mm penalty per unit of source-count mismatch before normalization by the true source count. Because MxNE returned only one source in 35 trials, the reported reduction reflects both spatial localization and source-count performance. In separate sensitivity experiments, QUBO remained competitive across the tested sensor-noise and source-count conditions, although RAP-MUSIC performed comparably to or better than QUBO in some low-noise and three-source settings. The present experiments use classical simulated annealing and do not evaluate quantum hardware or claim quantum advantage. The resulting binary quadratic objective admits a direct Ising representation, enabling future evaluation on quantum-annealing and hybrid backends.

Electrodermal activity (EDA) signals, which reflect sympathetic nervous system arousal through changes in skin conductance, are widely used in psychological and behavioral research. Decomposing an observed EDA signal into its slowly varying tonic baseline and stimulus-driven phasic component is an important preprocessing step; however, existing methods process signals in isolation and remain highly sensitive to noise and motion artifacts. This work introduces gmsEDA, a new decomposition method based on generalized matrix separation whose model is designed to cope with noise and motion artifacts. Our method analyzes multiple recordings jointly rather than one at a time, taking advantage of patterns shared across signals to produce more accurate and robust results. Numerical experiments on both simulated and real data shows that this approach outperforms existing standard tools.

Spoken dialogue is a natural form of human–computer interaction, yet most speech language models remain limited to turn-based operation and lack real-time adaptability, such as user barge-in. Recent duplex speech-to-speech and speech-to-text models reduce latency by replacing multi-stage pipelines, but often compromise speech quality because accurate ASR, interruption handling, and high-fidelity synthesis must be optimized jointly. We propose VoiceChat-TTS, a low-latency, continuous, and streamable text-to-speech model for interactive agents. VoiceChat-TTS is driven directly by LLM text-token streams, supports explicit interruption via control tokens, and produces silence when no textual input is available. The model enables always-on, responsive speech generation while preserving modularity and high speech quality, and it supports mid-utterance interruptions without resetting the KV cache.

In parallel to the cyber attack that manipulates the reference points of distributed energy resources (DERs) by maliciously accessing the remote monitoring and control system, the vulnerability of voltage/current sensors to electromagnetic interference (EMI) in the physical domain has been widely discussed. Existing research efforts against sensor spoofing attacks can be classified into physical prevention and cyber detection/mitigation. These defence methods each have strengths and weaknesses in balancing cost, security, and performance in a single DER, yet systematic research on their multi-layer efficient coordination across DERs remains limited. Towards this end, this paper proposes a hierarchical framework to detect and mitigate sensor spoofing attacks in networked microgrids (NMGs) via {multi-layer cyber-physical coordination}. It requires only to deploy physical prevention technologies at critical points, i.e., the local points of common coupling (PCC) of MGs, such that cyber detection/mitigation algorithms can be adopted based on the secured sensor readings to counter sensor spoofing attacks in DERs. The framework employs an MG-DER coordinated proactive detection scheme to strategically trigger parameter perturbations, under which the intelligent sensor spoofing attacks can be {effectively} disclosed. Afterwards, mitigation schemes based on MG-DER coordination are activated to recursively and accurately estimate sensor biases. Experiments on a cyber-physical DC NMG testbed confirm the framework's effectiveness across diverse attack scenarios.

This paper studies safety analysis and controller synthesis for partially observable nonlinear control systems. We extend the control barrier–value function (CBVF) framework, which combines Hamilton–Jacobi reachability and control barrier functions, to settings where full state information is not available and control is based on an estimated state. Given an estimator, we apply conformal prediction to the estimation error and obtain an error bound at a user-chosen miscoverage level. We incorporate this bound into the estimator-space safety analysis and define a CBVF-based safety certificate for partially observable systems. We then derive a finite-horizon probabilistic safety guarantee for the true system state. Finally, we propose a QP-based online safety filter for systems affine in the control and disturbance, whose solution enforces the CBVF safety condition in real time against bounded disturbance. The proposed framework is illustrated on a partially observable obstacle-avoidance case study.

Human speech production is constrained by physiology, giving rise to characteristic temporal structure on acoustic signals. We hypothesise that these constraints manifest as structured trajectory dynamics in the latent space of Self-Supervised Learning (SSL) models, and that synthetic speech violates them detectably. To test this hypothesis, we train a causal Long Short-Term Memory (LSTM) next-frame predictor on bonafide speech only (Stage 1), using the deepfake-specialised SSL backbone Wav2Vec2-Large-AntiDeepfake, and compare against a static global-average-pooling baseline using identical features, thus isolating the contribution of temporal modelling. A supervised Stage 2, which trains a Multi-Layer Perceptron on the frozen LSTM internal states using labelled data, is included to characterise the role of spoof supervision. Our system achieves competitive or state-of-the-art performance across six benchmarks: ASVspoof 2019/2021, Codecfake, In-the-Wild, MLAAD-EN, and Deepfake-Eval-2024, including best published EER on ASVspoof 2021 (0.75%) and, notably, Stage 1 trained on bonafide speech only surpasses the published supervised baseline from the same backbone on DE2024 (30.35%). On near-domain benchmarks, static and dynamic approaches perform comparably. On harder cross-corpus benchmarks with diverse synthesis methods, trajectory dynamics provide substantial gains, confirming that temporal physiological constraints carry detection signal beyond utterance-level statistics.

The rapid proliferation of multi-standard 5G New Radio sub-6 GHz systems have intensified the demand for compact, high-isolation multiplexing components capable of simultaneously routing multiple frequency channels through a single shared antenna port without cross-band interference. This paper presents the design and full-wave electromagnetic simulation of a compact microstrip triplexer operating at 2.2, 2.6, and 3.0 GHz for 5G sub-6 GHz applications. The proposed triplexer employs square open-loop resonators arranged as three independent three-pole Chebyshev bandpass filter channels, yielding nine resonator poles in total. Each channel is synthesised from the standard normalised Chebyshev lowpass prototype. The three channels are integrated at a common input port via a T-junction, with connecting transmission line stubs dimensioned to enforce high inter-channel isolation. Full-wave EM simulation results demonstrate return losses of 21.1 dB, 23.1 dB, and 22.8 dB; insertion losses of 1.08 dB, 1.01 dB, and 0.98 dB; and inter-channel isolations of 45.7 dB, 45.2 dB, and 45.7 dB, respectively. The achieved inter-channel isolation exceeding 45 dB represents a significant improvement over the majority of recently reported microstrip triplexer designs.

Positron emission tomography (PET) imaging suffers from limited spatial resolution and low signal-to-noise ratio, which can compromise quantitative accuracy and lesion detectability. Deep learning-based denoising methods have demonstrated strong potential for improving PET image quality. However, their practical deployment in real-world settings remains challenging, often requiring multiple specialized models and expert interventions, such as identifying motion-induced misregistration artifacts, estimating noise levels to select an appropriate denoiser, and performing lesion-focused quantitative assessment after denoising. Recent advances in vision-language models (VLMs) for image quality understanding and large language models (LLMs) for contextual reasoning provide new opportunities for automated, decision-driven workflows. Inspired by expert workflows for PET image quality enhancement, we propose an VLM- and LLM-driven multi-agent PET denoising framework that dynamically assesses image quality and lesion status, autonomously selects optimal denoising models and parameters, and enables closed-loop feedback with rollback mechanisms. Experiments were conducted on Siemens Biograph Vision Quadra PET/CT data with 1/20 and 1/50 low-dose settings. Individual module evaluations demonstrated the reliability of the agentic components, while the complete framework achieved higher PSNR and SSIM than UNet, GAN, and DDPM baselines at both dose levels. These preliminary results demonstrate the feasibility of using a closed-loop multi-agent framework to adapt PET denoising strategies to different image conditions.

This paper studies energy-efficient downlink multi-user transmissions with unmanned aerial vehicle (UAV) communication systems equipped with stacked intelligent metasurfaces (SIM), enabling wave-domain analog beamforming through multiple cascaded metasurface layers, while low-dimensional digital precoding is carried out using a limited number of transmit radio-frequency chains. This architecture enables flexible electromagnetic wave manipulation with reduced hardware complexity, making it particularly suitable for energy-constrained aerial platforms. We formulate a hardware-aware energy-efficiency (EE) maximization problem aiming to jointly optimize the digital precoder, the phase shifts of all SIM layers, and the three-dimensional UAV position under transmit-power, SIM operation, and UAV deployment constraints. The resulting problem is highly non-convex due to the fractional objective, the cascaded SIM structure and the unit-modulus phase constraints of the constituent metasurface layers, as well as the non-linear UAV-dependent channel. To address these challenges, we develop a transform-based alternating optimization framework that combines Dinkelbach's method, dual and quadratic transforms, to enable closed-form digital beamforming, Riemannian manifold optimization for SIM phase shifts, and successive convex approximation (SCA) for UAV positioning. Convergence and complexity analyses are provided to characterize the proposed algorithm. The presented numerical results showcase that the proposed joint design significantly improves EE compared with fully digital and maximum ratio transmission benchmark schemes, while revealing important design trade-offs among transmit power, SIM size, and the number of its constituent stacked layers.

Streaming automatic speech recognition (ASR) underperforms on domain-shifted target audio, where labeled in-domain data is costly to prepare while unlabeled audio is abundant. We present StreamHear, a semi-supervised pipeline that adapts a pretrained streaming student by fine-tuning an offline transducer teacher on the labeled training set, generating pseudo-labels on the unlabeled portion, and fine-tuning the student on the mixture. We further introduce a prior-regularized dynamic-programming realignment step that fixes chunk-level word placement using an ASR-hypothesis anchor. Across four datasets spanning financial calls, prepared read speech, and phone-quality dialogue, StreamHear consistently outperforms supervised student fine-tuning and narrows the gap to the offline teacher.

Computer-aided detection (CADe) systems for colonoscopy promise to reduce clinical miss rates, yet reliable real-world deployment remains elusive. This translational gap stems in part from a structural flaw in model development: the reliance on curated datasets that under-represent the long negative stretches and procedure-related artifacts characteristic of routine examinations. Training and evaluating architectures strictly on these lesion-centric benchmarks creates an illusion of success, since such benchmarks cannot capture clinically crucial metrics. To expose this gap, we establish TRUE-Colon, a standardized benchmarking protocol that measures key deployment characteristics alongside localization accuracy, and evaluate four real-time architectures (Faster R-CNN, YOLOv8, YOLOv11, RT-DETR) across curated benchmarks (SUN, PICCOLO) and 60 unedited, full-length procedures (REAL-Colon). We observe a consistent transfer asymmetry: models trained strictly on curated clips suffer a severe performance collapse when evaluated on full procedures, whereas procedure-trained models substantially improve rejection of non-polyp content on REAL-Colon, and largely retain their accuracy on curated benchmarks. Beyond transferability, we find that the Transformer detector attains the strongest sensitivity and the earliest, most persistent detections, while the convolutional detectors stay competitive at a higher throughput. Together, these results indicate that both training and benchmarking for deployable CADe should shift from curated, lesion-centric clips toward full-procedure data and deployment-relevant operating points. Source code is available at https://github.com/sdoerrich97/true-colon.

We solve exactly a fundamental problem of adaptive control against adversarial disturbances: regulate the scalar system $x_{t+1} = ax_t + u_t + w_t$, $x_0=0$, $\|w\|_\infty \le 1$, where the constant pole $a \in [-\Delta, \Delta]$ is unknown in sign and magnitude and $\Delta$ is arbitrarily large. Elementary as the system looks, the least worst-case peak $\|x\|_\infty$ that a causal controller can guarantee against an adversarial pair $(a, w)$ (the value of this game) has, to our knowledge, never been determined for any adaptive control problem with parametric uncertainty of arbitrary size under this criterion; existing theory supplies stability certificates, gain bounds, and regret rates, not the value. That value is $\gamma^\star(\Delta) = 1 + \Delta$ for every $\Delta>0$. The summand $1$ is the irreducible price of the disturbance, and $\Delta$ the exact price of a single, unavoidable identification spike. The optimal policy is certainty-equivalent deadbeat control at the midpoint of the set-membership consistent interval, an instance of the robust oracle $\times$ consistent model chasing architecture. The architecture is forced, not merely sufficient: writing $\theta_t := -u_t/x_t$ exhibits every causal controller as an oracle-selector composition, and optimality pins the selector to the midpoint at the critical histories. The standard tools, classical and modern, each fail quantifiably: probing is punished before it pays, commitment is fatal at sub-disturbance excitation once adaptation is necessary, optimism degenerates to tie-breaking or pays asymptotically at least twice the optimum, and regret certificates are blind to the worst-case peak in both directions. The optimal law contains no exploration mechanism, its learning purely passive. These results give the first exact optimality certificate for consistent model chasing as a design principle for adversarial adaptive control.

In this paper, we formalize a joint communication-control strategy optimization (JCCO) problem in multi-agent linear systems with quadratic costs, under the common-information-based (CIB) framework from decentralized stochastic control. For computational tractability, we focus on such JCCO problems with partially nested (PN) information structures (ISs). In particular, with a baseline communication protocol that leads to a PN IS, we establish a series of conditions under which the partial nestedness is preserved under the (additional) communication strategies to be optimized, while violating them may cause nonlinearity of the optimal strategies in general, with open-loop communication strategies. We then develop a dynamic-programming-based approach to compute the optimal control strategies of JCCO with open-loop communication strategies, which yields a set of closed-form Riccati Equations. As a byproduct of independent interest, such an approach also offers a way to solve decentralized linear-quadratic control with PN ISs and output feedback, under the CIB framework. Finally, we extend such an approach to JCCOs with closed-loop communication strategies, yielding a more tractable dynamic program than an infinite-dimensional CIB-belief-based one.

We study safety-critical control for teams of quadrotor UAVs driven by decentralized navigation functions under learned model uncertainty. These functions generate fully actuated translational reference forces, while quadrotors can only produce thrust along their body-fixed vertical axes. We construct a thrust-attitude implementation of the induced navigation forces and quantify its error with respect to the fully actuated reference dynamics. An aggregated robust HOCBF-QP safety filter minimally modifies the nominal thrusts while guaranteeing pairwise collision avoidance with high probability.

Complex multipath environments are usually avoided in wave-based information processing because repeated scattering creates many interfering propagation paths, obscuring controllability and generating extreme sensitivity to perturbations. The addition of nonlinear mechanisms fundamentally alters the wave-control landscape by breaking the superposition principle that underpins most wave-management strategies. Here, we show that these two apparent impediments – multipath complexity and nonlinearity – can instead be harnessed as key resources for physical optimization. We demonstrate an in-situ adjoint optimization protocol in a wave-chaotic platform incorporating a single localized nonlinear defect, in which the system itself performs both the forward and the adjoint propagations required for gradient evaluation. Recurrent multipath returns repeatedly expose the wave to the defect, producing from a minimal hardware a rich nonlinear input-output map with many pathway-mediated degrees of freedom. At the same time, a suitable adjoint excitation enables direct extraction of the sensitivities from measurements alone, without a digital twin or conventional numerical backpropagation. We experimentally validate the protocol on a minimal nonlinear multipath platform composed of incommensurate coaxial cables connected via T-junctions, one of which hosts a diode-loaded cavity. Our approach opens a route to adaptive wireless communications, imaging and analog intelligence in complex, partially unknown environments where conventional modeling is impractical.

Analog circuit design is a time-consuming, iterative process in a nonlinear and high-dimensional design space that relies heavily on expert intuition. Among recent developments, LLMs have introduced a promising approach by bringing natural language reasoning to circuit design tasks. The majority of conventional LLM-based approaches provide fragmented solutions that focus either only on sizing or topology generation. These methods require adding specific technical knowledge manually, which is inefficient and prone to hallucinations during circuit sizing. Moreover, the inherent trade-off in meeting different specs makes current approaches iterative and tedious. Another shortcoming is the inability to create innovative topologies, which may lead to sub-optimal designs due to reliance on conventional topologies. In this paper, we present AaLLM, an open-source end-to-end multi-agent LLM workflow that takes user specs as input and outputs the appropriate netlist, encompassing both topology generation and circuit sizing. AaLLM automates the creation of a relevant knowledge base from research papers and textbooks to combat tedious manual data collection. A RAG model is implemented to emulate circuit design expertise using this knowledge base. Moreover, AaLLM uses a novel tri-agent feedback system comprising a Designer that determines circuit component values, a Critic that scrutinizes these values, and an Evaluator that minimizes circuit sizing iterations by arbitrating between the other two agents. AaLLM-generated novel topologies achieve a figure of merit (FoM) comparable to that of known topologies, and up to 3x higher for certain circuits. Testing on several circuit topologies, our results show a 3x - 4.5x decrease in the number of SPICE calls at inference when compared to SOTA multi-agent LLM pipelines. The results also show a 40x decrease in wall-clock time compared to existing approaches.

This letter proposes the integration of on-off digital noise (OODN) modulation with fluid antenna systems (FASs). A unified analytical framework is developed to evaluate the performance of FAS- assisted OODN receivers over additive white Gaussian noise and generalized ąppa}-{\mu} fading channels. The analysis incorporates fluid antenna port selection, the number of available ports, and spatial correlation. Analytical expressions for the average bit error probability are derived and the achievable diversity order is characterized. All expressions are validated through Monte Carlo simulations. Results demonstrate that the spatial diversity provided by the FAS significantly enhances the reliability of OODN transmissions while preserving their inherent low-complexity, non-coherent operation without carrier-phase recovery. The proposed framework establishes a new research direction for energy-efficient Internet of Things (IoT) and machine-type communication systems.

This paper proposes a parameter estimation scheme for uplink MIMO systems assisted by a Hybrid Reconfigurable Intelligent Surface (HRIS). To reduce hardware complexity, the HRIS employs a small number of active elements arranged in a sparse coprime geometry for local sensing, while the remaining elements passively reflect signals. This simultaneous sensing and reflection significantly improves accuracy over fully passive architectures. The method leverages Khatri-Rao and Kronecker factorizations to efficiently decouple the cascaded channel at the base station. Furthermore, spatial smoothing resolves the coprime array rank deficiency, enabling robust angular extraction via Root-MUSIC. Simulations demonstrate highly resilient estimation performance across scenarios.

Self-supervised learning (SSL) speech representations achieve strong performance for Parkinson's disease (PD) detection within individual corpora. However, it remains unclear whether these models capture disease-related characteristics or exploit dataset-specific confounds, particularly since most SSL backbones are pretrained exclusively on healthy speech. To investigate this question, we perform a layer-wise analysis of nine SSL speech backbones using a low-capacity logistic regression probe across three languages. We structure the evaluation as multiple scenarios that progressively introduce distribution shifts in participant identity, recording conditions, language, and pathology. Our results reveal two key findings. First, layer selection is highly corpus-dependent: the optimal representation layer is determined primarily by the source dataset rather than by the SSL architecture itself. Second, the transferred discriminative signal lacks pathological specificity: classifiers trained to detect PD assign similarly high probabilities to both PD and dementia speech in the target corpus. These results highlight critical limitations that must be addressed before speech-based pathology recognition models can be reliably deployed in clinical settings.

6G networks will not be serving as communication infrastructures only; rather, they are expected to evolve into intelligent systems, where thousands of autonomous artificial intelligence (AI) agents are interconnected. The agents are deployed across a wide range of platforms including low Earth orbit (LEO) satellites, high-altitude platforms (HAPs), unmanned aerial vehicles (UAVs), edge servers, and terrestrial devices. These agents continuously observe their environment and exchange information. Semantic communication provides an efficient mechanism for exchanging meaningful information instead of raw data. However, its effectiveness depends on the communicating agents having sufficiently aligned beliefs to correctly interpret and decode the transmitted messages. This assumption becomes difficult to satisfy in the 6G network where heterogeneous AI models operate under diverse computational constraints and continuously acquire different knowledge from their local environments. This article presents a heterogeneity-aware belief synchronization framework for 6G AI-native networks. It uses latent translation models deployed on multi-access edge computing (MEC) servers. These models translate belief updates from one agent to agent-specific knowledge without requiring joint training and a homogeneous architecture of models. By exchanging compact belief updates through a latent translation model only when necessary, the framework preserves privacy, reduces synchronization cost, and minimizes local knowledge drift. We validate the framework through a case study on a multi-layered terrestrial/non-terrestrial network. Results demonstrate that it maintains low synchronization cost, measured by the number of parameters transmitted, and low belief alignment error across the heterogeneous agents in the case study.

Wireless networks increasingly support closed-loop industrial applications in which sensed data must be transmitted, processed, and converted into an action before physical process makes the result obsolete. Throughput, latency, and inference accuracy measured separately cannot determine whether such an application completed a useful task. We propose a Process-Aware Co-adaptation Engine framework that combines application outcomes, process state, radio telemetry, edge-compute state, and sensing configuration to select coordinated operating points across the complete loop. We evaluate the proposed approach in a factory-inspection case study that integrates a physics-based digital twin, a programmable 5G O-RAN network, and edge-based visual inference. The experiments show that the preferred resource allocation changes with production speed and that adapting individual system components independently can be inefficient. We further show that efficient configurations can be identified with relatively few full-system evaluations.

The space-filling input design approach generates a so-called space-filling dataset in the feature space of the system model. The design method is applicable on a broad class of model structures with wide selection of signals and also incorporates information measures through optimality criteria into the signal design. However, during the signal design, knowledge of a hypothesized model is required. The designed signal can perform far from the optimal if the true system is significantly different from the hypothesized system model. This paper proposes a robust space-filling input design algorithm that can generate a space-filling dataset for an entire class of models. The proposed algorithm takes the expectation of an optimality criterion over the population of the model class, and a stochastic approximation technique is employed to optimize this robust criteria. The efficiency of the proposed algorithm is demonstrated in a simulation example.

Recent work has explored elaborate beamfocusing techniques for the radiative nearfield, with some studies suggesting that certain beamshapes, such as Airy beams, can enable efficient electromagnetic (EM) wave transmission behind obstacles. In this letter, we ask whether the added complexity of such techniques is justified. We distinguish between partially and fully occluded regions. In the partially occluded region, where Airy beams are commonly employed, we show that a simple line-of-sight (LoS) strategy, which activates only antennas having an unobstructed view of the receiver, is near-optimal and outperforms Airy beams at substantially lower complexity. In the fully occluded region, we argue that accurate beamfocusing requires a physically consistent EM wave propagation model that captures propagation effects such as diffraction. Once such a model is available, however, the optimal beamfocusing strategy has a closed-form solution and can be computed directly. These results suggest that elaborate techniques, such as Airy beams, offer little benefit over simpler alternatives for beamfocusing into occluded regions.

Foundation models (FMs) trained on large-scale accelerometer data have been proposed as general-purpose feature extractors for health monitoring, but systematic evidence of their advantages is lacking. We present the first comprehensive evaluation of four open-source accelerometer FMs against supervised baselines covering 19 tasks across the domains of activity recognition including activities of daily living, clinical monitoring, and physiological inference. We find task-dependent performance results: supervised models remain competitive with FMs on human action recognition (HAR), with no consistent advantage for either, while selected FMs lead on fall and stress detection and are the most robust to sensor-placement variation. As frozen feature extractors, FMs are strongest for demographic inference, whereas sleep staging performance remains near chance level for all models. The internal FM representations show strong similarity across layers, highlighting potential for future FM improvements. Linear and frozen probing reveals that UniMTS provides the strongest representations and is the only FM that surpasses the supervised baselines without finetuning. Concept discovery analysis shows all models capture high-intensity activities clearly but struggle with sedentary, complex or ambiguous activities. We provide scenario-based deployment recommendations. Furthermore, we identify FM-derived activity profile inference-moving beyond fixed category classification-as a promising research direction.