Papers

Confounding pathology with normal anatomical variation remains a significant challenge in unsupervised medical-image anomaly detection, resulting in numerous false positives. To enhance integration of healthy variation, we augment the latent representation of a CNN autoencoder with contextual similarities within a normal cohort through batch-wise hypergraph estimation and a shared-weights graph convolution layer, producing a population-aware embedding. On a heterogeneous brain-tumor dataset of 2D MRI scans, the method improves separability between healthy and pathological samples, achieving an AUC-ROC of 0.90 (95% CI 0.84-0.95, 5.7% absolute gain), and a 16% absolute improvement in average precision (0.78 AP, 95% CI 0.66-0.89), thereby lowering false-positive rates. Moreover, both anomaly detection and downstream tumor versus no-tumor classification performance improve with the size of the mini-batch context captured in the augmented representation, suggesting a tunable lever for integrating healthy variation.

This paper presents a scenario generation framework that creates diverse, parametrized, and safety-critical driving situations to validate the safety features of autonomous vehicles in simulation [15]. By modeling factors such as road geometry, traffic participants, environmental conditions, and perception uncertainties, the framework enables repeatable and scalable testing of safety mechanisms, including emergency braking, evasive maneuvers, and vulnerable road user protection. The framework supports both regulatory and edge case scenarios, mapped to hazards and safety goals derived from Hazard Analysis and Risk Assessment (HARA), ensuring traceability to ISO 26262 functional safety requirements and performance limitations. The output from these simulations provides quantitative safety metrics such as time-to-collision, minimum distance, braking and steering performance, and residual collision severity. These metrics enable the systematic evaluation of evasive maneuvering as a safety feature, while highlighting system limitations and edge-case vulnerabilities. Integration of scenario-based simulation with safety engineering principles offers accelerated validation cycles, improved test coverage at reduced cost, and stronger evidence for regulatory and stakeholder confidence.

Sensing is an integral part of 6G and beyond systems, providing exceptional environmental perception along with communication. Radio frequency (RF)-based sensing often relies on simplified geometric assumptions (e.g., point scatterers or planar surfaces) to model specular multipath and keep inference tractable. However, such representations are limited in their ability to capture extended objects with complex geometries and properties. This paper presents a probabilistic occupancy grid framework for radio-based simultaneous localization and mapping (SLAM), jointly reconstructing geometric structures and their RF-related properties. The proposed occupancy grid map representation is integrated into a multipath-based SLAM formulation to enable simultaneous mobile-agent localization and environment mapping using multipath measurements. To connect RF measurements with the grid map, a surface model is employed to describe candidate reflection paths, while occupancy grid cell states capture measurement uncertainties and fine-grained geometric details. RF-related object properties are represented through reflection coefficients. The proposed framework offers a principled, proof-of-concept approach to physically interpretable radio-based mapping, and simulation results demonstrate accurate reconstruction of geometry and material properties, as well as high-accuracy localization. In addition, the results highlight the potential to use prior occupancy maps obtained from other radio devices or complementary sensors for subsequent map extension and refinement.

Reliable positioning in dense urban environments remains challenging due to frequent GNSS signal blockage, multipath, and rapidly varying satellite geometry. While factor graph optimization (FGO)-based GNSS-IMU fusion has demonstrated strong robustness and accuracy, most formulations remain offline. In this work, we present a real-time tightly coupled GNSS-IMU FGO method that enables causal state estimation via incremental optimization with fixed-lag marginalization, and we evaluate its performance in a highly urbanized GNSS-degraded environment using the UrbanNav dataset.

Accurate positioning, navigation, and timing (PNT) is fundamental to the operation of modern technologies and a key enabler of autonomous systems. A very important component of PNT is the Global Navigation Satellite System (GNSS) which ensures outdoor positioning. Modern research directions have pushed the performance of GNSS localization to new heights by fusing GNSS measurements with other sensory information, mainly measurements from Inertial Measurement Units (IMU). In this paper, we propose a loosely coupled architecture to integrate GNSS and IMU measurements using a Factor Graph Optimization (FGO) framework. Because the FGO method can be computationally challenging and often used as a post-processing method, our focus is on assessing its localization accuracy and service availability while operating in real-time in challenging environments (urban canyons). Experimental results on the UrbanNav-HK-MediumUrban-1 dataset show that the proposed approach achieves real-time operation and increased service availability compared to batch FGO methods. While this improvement comes at the cost of reduced positioning accuracy, the paper provides a detailed analysis of the trade-offs between accuracy, availability, and computational efficiency that characterize real-time FGO-based GNSS/IMU fusion.

This paper examines the problem of achieving "grace" when controlling dynamical systems for safety, which is defined in terms of providing multi-layered safety assurances. Namely, two safety layers are created: a primary layer that represents a desirable degree of safety, and a secondary failsafe layer. Graceful control then involves ensuring that even if the primary layer is breached, the failsafe layer remains forward invariant. The paper pursues this goal by constructing a safety constraint that combines the concepts of zeroing and reciprocal control barrier functions with regard to the primary and secondary safe sets, respectively. This constraint is analogous to a stiffening spring, making it possible to construct energy-based analytical proofs of the resulting graceful safety guarantees. The proposed approach is developed for systems with a relative degree of either 1 or 2, the latter case being particularly useful for mechanical systems. We demonstrate the applicability of the method using a wall collision avoidance example. This demonstration highlights the benefits of the proposed approach compared to traditional benchmarks from the literature.

Cyber-security systems often operate in resource-constrained environments, such as edge environments and real-time monitoring systems, where model size and inference time are crucial. A light-weight intrusion detection framework is proposed that utilizes the Kolmogorov-Arnold Network (KAN) to capture complex features in the data, with the efficiency of decoupled knowledge distillation (DKD) training approach. A high-capacity KAN network is first trained to detect attacks performed on the test bed. This model then serves as a teacher to guide a much smaller multilayer perceptron (MLP) student model via DKD. The resulting DKD-MLP model contains only 2,522 and 1,622 parameters for WADI and SWaT datasets, which are significantly smaller than the number of parameters of the KAN teacher model. This is highly appropriate for deployment in resource-constrained devices with limited computational resources. Despite its low size, the student model maintains a high performance. Our approach demonstrate the practicality of using KAN as a knowledge-rich teacher to train much smaller student models, without considerable drop in accuracy in intrusion detection frameworks. We have validated our approach on two publicly available datasets. We report F1-score improvements of 4.18% on WADI and 3.07% on SWaT when using the DKD-MLP model, compared to the bare student model. The implementation of this paper is available on our GitHub repository.

Low-thrust electric propulsion missions are often designed under simplifying assumptions such as constant thrust or fixed specific impulse, neglecting the strong coupling between trajectory dynamics, spacecraft power availability, and propulsion performance. In deep-space environments with reduced solar irradiance, these assumptions can lead to suboptimal or infeasible designs, underscoring the need to simultaneously optimize the trajectory and power subsystem. This paper presents a multidisciplinary design optimization (MDO) framework for the simultaneous design of low-thrust trajectories and spacecraft power systems, with explicit coupling to electric propulsion performance. The framework incorporates a high-fidelity variable-specific impulse model of the SPT-140 Hall thruster, in which thrust and efficiency are directly constrained by time-varying solar power availability and solar array degradation, rather than treated as fixed parameters. The coupled problem is posed as a time-optimal control problem and addressed using a framework built on top of OpenMDAO and Dymos toolchains, where Dymos employs a collocation-based direct-transcription approach for trajectory optimization. OpenMDAO provides accurate analytic partial derivatives, enabling efficient gradient-based optimization. A Fast Fourier Series shape-based method is used to generate dynamically feasible initial guess trajectories, and the resulting nonlinear programming problem is solved using IPOPT. The proposed framework is demonstrated through a low-thrust orbit insertion scenario around asteroid 16-Psyche, a regime in which reduced solar irradiance makes power-aware trajectory design particularly critical. Simulation results demonstrate the framework's ability to capture key power-propulsion-trajectory trade-offs, highlighting the importance of integrated power optimization for realistic electric propulsion mission design.

Early and accessible detection of Alzheimer's disease (AD) remains a major challenge, as current diagnostic methods often rely on costly and invasive biomarkers. Speech and language analysis has emerged as a promising non-invasive and scalable approach to detecting cognitive impairment, but research in this area is hindered by the lack of publicly available datasets, especially for languages other than English. This paper introduces the PARLO Dementia Corpus (PDC), a new multi-center, clinically validated German resource for AD collected across nine academic memory clinics in Germany. The dataset comprises speech recordings from individuals with AD-related mild cognitive impairment and mild to moderate dementia, as well as cognitively healthy controls. Speech was elicited using a standardized test battery of eight neuropsychological tasks, including confrontation naming, verbal fluency, word repetition, picture description, story reading, and recall tasks. In addition to audio recordings, the dataset includes manually verified transcriptions and detailed demographic, clinical, and biomarker metadata. Baseline experiments on ASR benchmarking, automated test evaluation, and LLM-based classification illustrate the feasibility of automatic, speech-based cognitive assessment and highlight the diagnostic value of recall-driven speech production. The PDC thus establishes the first publicly available German benchmark for multi-modal and cross-lingual research on neurodegenerative diseases.

Many essential manipulation tasks - such as food preparation, surgery, and craftsmanship - remain intractable for autonomous robots. These tasks are characterized not only by contact-rich, force-sensitive dynamics, but also by their "implicit" success criteria: unlike pick-and-place, task quality in these domains is continuous and subjective (e.g. how well a potato is peeled), making quantitative evaluation and reward engineering difficult. We present a learning framework for such tasks, using peeling with a knife as a representative example. Our approach follows a two-stage pipeline: first, we learn a robust initial policy via force-aware data collection and imitation learning, enabling generalization across object variations; second, we refine the policy through preference-based finetuning using a learned reward model that combines quantitative task metrics with qualitative human feedback, aligning policy behavior with human notions of task quality. Using only 50-200 peeling trajectories, our system achieves over 90% average success rates on challenging produce including cucumbers, apples, and potatoes, with performance improving by up to 40% through preference-based finetuning. Remarkably, policies trained on a single produce category exhibit strong zero-shot generalization to unseen in-category instances and to out-of-distribution produce from different categories while maintaining over 90% success rates.

The shock response spectrum (SRS) is widely used to characterize the response of single-degree-of-freedom (SDOF) systems to transient accelerations. Because the mapping from acceleration time history to SRS is nonlinear and many-to-one, reconstructing time-domain signals from a target spectrum is inherently ill-posed. Conventional approaches address this problem through iterative optimization, typically representing signals as sums of exponentially decayed sinusoids, but these methods are computationally expensive and constrained by predefined basis functions. We propose a conditional variational autoencoder (CVAE) that learns a data-driven inverse mapping from SRS to acceleration time series. Once trained, the model generates signals consistent with prescribed target spectra without requiring iterative optimization. Experiments demonstrate improved spectral fidelity relative to classical techniques, strong generalization to unseen spectra, and inference speeds three to six orders of magnitude faster. These results establish deep generative modeling as a scalable and efficient approach for inverse SRS reconstruction.

Deep generative models have become a standard for modeling priors for inverse problems, going beyond classical sparsity-based methods. However, existing theoretical guarantees are mostly confined to finite-dimensional vector spaces, creating a gap when the physical signals are modeled as functions in Hilbert spaces. This work presents a rigorous framework for generative compressed sensing in Hilbert spaces. We extend the notion of local coherence in an infinite-dimensional setting, to derive optimal, resolution-independent sampling distributions. Thanks to a generalization of the Restricted Isometry Property, we show that stable recovery holds when the number of measurements is proportional to the prior's intrinsic dimension (up to logarithmic factors), independent of the ambient dimension. Finally, numerical experiments on the Darcy flow equation validate our theoretical findings and demonstrate that in severely undersampled regimes, employing lower-resolution generators acts as an implicit regularizer, improving reconstruction stability.

A framework of continuous-aperture array (CAPA)-based integrated sensing and communications (ISAC) under a fading communication channel is proposed. A continuous operator-based signal model is developed, and the statistics of the communication channel gain are characterized via Landau's eigenvalue theorem. On this basis, the performance of the CAPA-based ISAC system is analyzed by considering three continuous beamforming designs: i) the sensing-centric (S-C) design that optimizes sensing performance, ii) the communication-centric (C-C) design that optimizes communication performance, and iii) the Pareto-optimal design that balances the sensing-communication trade-off. For the S-C and C-C design, closed-form expressions for the sensing rate (SR), ergodic communication rate (CR), and outage probability are derived, and high-signal-to-noise ratio asymptotic analysis is conducted to obtain the multiplexing and diversity gains. For the Pareto-optimal design, the Pareto-optimal beamformer achieving the Pareto boundary is derived, and the achievable SR-CR region is characterized. Numerical results demonstrate that the proposed CAPA-ISAC scheme outperforms both conventional spatially discrete arrays-based ISAC and CAPA-based frequency-division sensing and communications.

We present a framework that integrates EEG-based visual and motor imagery (VI/MI) with robotic control to enable real-time, intention-driven grasping and placement. Motivated by the promise of BCI-driven robotics to enhance human-robot interaction, this system bridges neural signals with physical control by deploying offline-pretrained decoders in a zero-shot manner within an online streaming pipeline. This establishes a dual-channel intent interface that translates visual intent into robotic actions, with VI identifying objects for grasping and MI determining placement poses, enabling intuitive control over both what to grasp and where to place. The system operates solely on EEG via a cue-free imagery protocol, achieving integration and online validation. Implemented on a Base robotic platform and evaluated across diverse scenarios, including occluded targets or varying participant postures, the system achieves online decoding accuracies of 40.23% (VI) and 62.59% (MI), with an end-to-end task success rate of 20.88%. These results demonstrate that high-level visual cognition can be decoded in real time and translated into executable robot commands, bridging the gap between neural signals and physical interaction, and validating the flexibility of a purely imagery-based BCI paradigm for practical human-robot collaboration.

No-regret learning dynamics ensure that a learner asymptotically achieves an average reward no worse than that of any fixed strategy. This no-regret guarantee does not determine the value of the asymptotic average reward. Indeed, it is possible for different no-regret learning dynamics to exhibit different asymptotic average rewards when facing the same environment while both assure the no-regret guarantee. This paper asks whether a "free-lunch" phenomenon can arise among no-regret algorithms. Namely, is it possible for one no-regret learning rule to uniformly outperform another no-regret learning rule across all payoff environments. Stated differently, can a learner regret not using a particular no-regret algorithm? We consider generalized replicator dynamics (RD) as a cascade interconnection between a linear time-invariant (LTI) system and the softmax nonlinearity. Varying this LTI system leads to different realizations of replicator dynamics, including so-called anticipatory RD, exponential RD, and other forms of higher-order RD. Setting the LTI system to be an integrator realizes standard RD, which is known to satisfy the no-regret property. Within this framework, we analyze and compare various realizations of these generalized realizations RD by varying the LTI system. We first formulate performance comparison as a passivity property of an associated comparison system and establish "local" dominance results, i.e., comparing the asymptotic performance near an equilibrium payoff vector. We then cast performance comparison between a form of anticipatory RD and standard RD as an optimal-control problem. We show that the minimal achievable cumulative reward gap is zero, thereby establishing global dominance of anticipatory RD across all payoff environments and establishing a "free lunch" among no-regret learning dynamics.

This paper presents a deep Q-network (DQN)-based gain-scheduling framework for safety-critical quadcopter trajectory tracking. Instead of directly learning control inputs, the proposed approach selects from a finite set of pre-certified stabilizing gain vectors, enabling reinforcement learning to operate within a structured and stability-preserving control architecture. By exploiting the isotropic structure of the translational dynamics, feedback gains are shared across spatial axes to reduce dimensionality while preserving performance. The learned policy adapts feedback aggressiveness in real time, applying high authority during large transients and reducing gains near convergence to limit control effort. Simulation results using a high-fidelity nonlinear quadcopter model demonstrate accurate trajectory tracking, bounded attitude excursions, smooth transition to hover after the final time, and consistent reward improvement, validating the effectiveness and robustness of the proposed learning-based gain scheduling strategy.

This research focuses on the design of Ka-band mobile antennas for satellite communication operating at 29 GHz. Starting from a single element and progressing to an 8x8 array, the antennas achieved a gain of up to 21 dB and return losses as low as -30 dB. The design process involves mathematical calculations and software implementation, utilizing parameters like patch dimensions, substrate properties, and effective permittivity. The chosen Ka-band frequency range, known for higher data transfer rates, addresses the demand for swift communication. Challenges in Ka-band mobile antenna design, including signal attenuation, directional accuracy, circular polarization, and impedance matching, are addressed through various configurations, including phased-array and electronically steerable antennas. This research focuses on the design of Ka-band mobile antennas for satellite communication at 29GHz, progressing from a single element to an 8x8 array. The antennas achieved a gain of up to 21 dB and return losses as low as -30 dB through mathematical calculations and software implementation using CST. Challenges in Ka-band antenna design, such as signal attenuation and impedance matching, are addressed through various configurations, including phased-array and electronically steerable antennas. Integration of machine learning techniques aids in optimization. In conclusion, this research advances high-frequency transmission technology, meeting the demands of modern satellite-based communication systems for applications like high-speed internet access and multimedia streaming. Keywords: Ka-band, mobile antennas, satellite communication, 29 GHz, antenna design, CST, high-speedinternet access, multimedia streaming

How do speech models trained through self-supervised learning structure their representations? Previous studies have looked at how information is encoded in feature vectors across different layers. But few studies have considered whether speech characteristics are captured within individual dimensions of SSL features. In this paper, we specifically look at speaker information using PCA on utterance-averaged representations. For a range of SSL models, we find that the principal dimension that explains most variance encodes pitch and associated characteristics like gender. Other individual principal dimensions correlate with intensity, noise levels, the second formant, and higher frequency characteristics. We then use synthesis analyses to show that the dimensions for most characteristics are isolated from each other's influence. We further show that characteristics can be changed by manipulating the corresponding dimensions.

Analyzing nonlinear systems with attracting robust invariant sets (RISs) requires estimating their domains of attraction (DOAs). Despite extensive research, accurately characterizing DOAs for general nonlinear systems remains challenging due to both theoretical and computational limitations, particularly in the presence of uncertainties and state constraints. In this paper, we propose a novel framework for the accurate estimation of safe (state-constrained) and robust DOAs for discrete-time nonlinear uncertain systems with continuous dynamics, open safe sets, compact disturbance sets, and uniformly locally $\ell_p$-stable compact RISs. The notion of uniform $\ell_p$ stability is quite general and encompasses, as special cases, uniform exponential and polynomial stability. The DOAs are characterized via newly introduced value functions defined on metric spaces of compact sets. We establish their fundamental mathematical properties and derive the associated Bellman-type (Zubov-type) functional equations. Building on this characterization, we develop a physics-informed neural network (NN) framework to learn the corresponding value functions by embedding the derived Bellman-type equations directly into the training process. To obtain certifiable estimates of the safe robust DOAs from the learned neural approximations, we further introduce a verification procedure that leverages existing formal verification tools. The effectiveness and applicability of the proposed methodology are demonstrated through four numerical examples involving nonlinear uncertain systems subject to state constraints, and its performance is compared with existing methods from the literature.

Semantic maps are increasingly utilized in areas such as robotics, autonomous systems, and extended reality, motivating the investigation of efficient compression methods that preserve structured semantic information. This paper studies lossless compression of semantic maps through a novel chain-coding-based framework that explicitly exploits contour topology and shared boundaries between adjacent semantic regions. We propose an extended chain code (ECC) to represent long-range contour transitions more compactly, while retaining a legacy three-orthogonal chain code (3OT) as a fallback mode for further efficiency. To efficiently encode sequences of ECC symbols, a context-adaptive entropy coding scheme based on Markov modeling is employed. Furthermore, a skip-coding mechanism is introduced to eliminate redundant representations of shared contours between adjacent semantic regions, supporting both complete and partial skips via run-length signaling. Experimental results demonstrate that the proposed method achieves an average bitrate reduction of 18% compared with a state-of-the-art benchmark on semantic map datasets. In addition, the proposed encoder and decoder achieve up to 98% and 50% runtime reduction, respectively, relative to a modern generic lossless codec. Extended evaluations on occupancy maps further confirm consistent compression gains across the majority of tested scenarios. The source code is made publicly available at https://github.com/InterDigitalInc/LosslessSegmentationMapCompression.

We present DLIOS, a Large Language Model (LLM)-augmented real-time multi-modal interactive enhancement overlay system for Douyin (TikTok) live streaming. DLIOS employs a three-layer transparent window architecture for independent rendering of danmaku (scrolling text), gift and like particle effects, and VIP entrance animations, built around an event-driven WebView2 capture pipeline and a thread-safe event bus. On top of this foundation we contribute an LLM broadcast automation framework comprising: (1) a per-song four-segment prompt scheduling system (T1 opening/transition, T2 empathy, T3 era story/production notes, T4 closing) that generates emotionally coherent radio-style commentary from lyric metadata; (2) a JSON-serializable RadioPersonaConfig schema supporting hot-swap multi-persona broadcasting; (3) a real-time danmaku quick-reaction engine with keyword routing to static urgent speech or LLM-generated empathetic responses; and (4) the Suwan Li AI singer-songwriter persona case study – over 100 AI-generated songs produced with Suno. A 36-hour stress test demonstrates: zero danmaku overlap, zero deadlock crashes, gift effect P95 latency <= 180 ms, LLM-to-TTS segment P95 latency <= 2.1 s, and TTS integrated loudness gain of 9.5 LUFS. live streaming; danmaku; large language model; prompt engineering; virtual persona; WebView2; WINMM; TTS; Suno; loudness normalization; real-time scheduling

In this work, we address the need for efficient and formally stable Recurrent Neural Networks (RNNs) in environments with limited computational resources by analyzing the stability of the Minimal Gated Unit (MGU) network, a lightweight alternative to common gated RNNs used in system identification. We derive sufficient parametric conditions for the MGU network's input-to-state stability and incremental input-to-state stability properties. These conditions enable a-posteriori validation of model stability and form the basis for novel stability-promoting training methodologies, including a warm-start of the network's parameters and a projected gradient-based optimization scheme, both of which are presented in this work. Comparative evaluation, including robustness analysis and validation on synthetic and real-world data (i.e., the Silverbox benchmark), demonstrates that the minimal gated unit network successfully combines formal stability guarantees with superior parameter efficiency and faster inference times compared to other state-of-the-art recurrent neural networks, while maintaining comparable and satisfactory accuracy. Notably, the results attained on the Silverbox benchmark illustrate that the stable MGU network effectively captures the system dynamics, whereas other stable RNNs fail to converge to a reliable model.

Resource allocation is a key factor in multiuser (MU) multiple-input multiple-output (MIMO) wireless systems to provide high quality of service to all user equipments (UEs). In congested scenarios, UE scheduling enables UEs to be distributed over time, frequency, or space in order to mitigate inter-UE interference. Many existing UE scheduling methods rely on greedy algorithms, which fail at treating the resource-allocation problem globally. In this work, we propose a UE scheduling framework for MU-MIMO wireless systems that approximately solves a nonconvex optimization problem that treats scheduling globally. Our UE scheduling framework determines subsets of UEs that should transmit simultaneously in a given resource slot and is flexible in the sense that it (i) supports a variety of objective functions (e.g., post-equalization mean squared error, capacity, and achievable sum rate) and (ii) enables precise control over the minimum and maximum number of resources the UEs should occupy. We demonstrate the efficacy of our UE scheduling framework for millimeter-wave massive MU-MIMO and sub-6-GHz cell-free massive MU-MIMO systems, and we show that it outperforms existing scheduling algorithms while approaching the performance of an exhaustive search.

This paper develops a nonlinear grid-forming (GFM) controller with provable voltage-formation guarantees, with over-current limiting enforced via a control-barrier-function (CBF)-based safety filter. The nominal controller follows a droop-based inner-outer architecture, in which voltage references and frequency are generated by droop laws, an outer-loop voltage controller produces current references using backstepping (BS), and an inner-loop current controller synthesizes the terminal voltage. The grid voltage is treated as an unknown bounded disturbance, without requiring knowledge of its bound, and the controller design does not rely on any network parameters beyond the point of common coupling (PCC). To robustify voltage formation against the grid voltage, a deadzone-adapted disturbance suppression (DADS) framework is incorporated, yielding practical voltage regulation characterized by asymptotic convergence of the PCC voltage errors to an assignably small and known residual set. Furthermore, the closed-loop system is proven to be globally well posed, with all physical and adaptive states bounded and voltage error transients (due to initial conditions) decaying exponentially at an assignable rate. On top of the nominal controller, hard over-current protection is achieved through a minimally invasive CBF-based safety filter that enforces strict current limits via a single-constraint quadratic program. The safety filter is compatible with any locally Lipschitz nominal controller. Rigorous analysis establishes forward invariance of the safe-current set and boundedness of all states under current limiting. Numerical results demonstrate improved transient performance and faster recovery during current-limiting events when the proposed DADS-BS controller is used as the nominal control law, compared with conventional PI-based GFM control.

Speech-based detection of cognitive impairment (CI) offers a promising non-invasive approach for early diagnosis, yet performance disparities across demographic and clinical subgroups remain underexplored, raising concerns around fairness and generalizability. This study presents a systematic bias analysis of acoustic-based CI and depression classification using the DementiaBank Pitt Corpus. We compare traditional acoustic features (MFCCs, eGeMAPS) with contextualized speech embeddings from Wav2Vec 2.0 (W2V2), and evaluate classification performance across gender, age, and depression-status subgroups. For CI detection, higher-layer W2V2 embeddings outperform baseline features (UAR up to 80.6%), but exhibit performance disparities; specifically, females and younger participants demonstrate lower discriminative power (\(AUC\): 0.769 and 0.746, respectively) and substantial specificity disparities (\(\Delta_{spec}\) up to 18% and 15%, respectively), leading to a higher risk of misclassifications than their counterparts. These disparities reflect representational biases, defined as systematic differences in model performance across demographic or clinical subgroups. Depression detection within CI subjects yields lower overall performance, with mild improvements from low and mid-level W2V2 layers. Cross-task generalization between CI and depression classification is limited, indicating that each task depends on distinct representations. These findings emphasize the need for fairness-aware model evaluation and subgroup-specific analysis in clinical speech applications, particularly in light of demographic and clinical heterogeneity in real-world applications.

Building speech deepfake detection models that are generalizable to unseen attacks remains a challenging problem. Although the field has shifted toward a pre-training and fine-tuning paradigm using speech foundation models, most approaches rely solely on supervised fine-tuning (SFT). Inspired by the field of large language models, wherein reinforcement learning (RL) is used for model fine-tuning, we investigate the impact of RL, specifically Group Relative Policy Optimization (GRPO). The results from experiments using multiple detectors and test sets indicate that pure GRPO-based fine-tuning improves performance on out-of-domain test sets while maintaining performance on target-domain test data. This approach outperforms both SFT-only and hybrid setups. Our ablation studies further suggest that the negative reward in GRPO may be a key factor in this improvement.

The performance of conventional speech enhancement systems degrades sharply in extremely low signal-to-noise ratio (SNR) environments where air-conduction (AC) microphones are overwhelmed by ambient noise. Although bone-conduction (BC) sensors offer complementary, noise-tolerant information, existing fusion approaches struggle to maintain consistent performance across a wide range of SNR conditions. To address this limitation, we propose the Deep Balanced Multimodal Iterative Fusion Framework (DBMIF), a three-branch architecture designed to reconstruct high-fidelity speech through rigorous cross-modal interaction. Specifically, grounded in a multi-scale interactive encoder-decoder backbone, the framework orchestrates an iterative attention module and a cross-branch gated module to facilitate adaptive weighting and bidirectional exchange. To complement this dynamic interaction, a balanced-interaction bottleneck is further integrated to learn a compact, stable fused representation. Extensive experiments demonstrate that DBMIF achieves competitive performance compared with recent unimodal and multimodal baselines in both speech quality and intelligibility across diverse noise types. In downstream ASR tasks, the proposed method reduces the character error rate by at least 2.5 percent compared to competing approaches. These results confirm that DBMIF effectively harnesses the robustness of BC speech while preserving the naturalness of AC speech, ensuring reliability in real-world scenarios. The source code is publicly available at github.com/wyl516w/dbmif.

Detecting speech from biosignals is gaining increasing attention due to the potential to develop human-computer interfaces that are noise-robust, privacy-preserving, and scalable for both clinical applications and daily use. However, most existing approaches remain limited by insufficient wearability and the lack of edge-processing capabilities, which are essential for minimally obtrusive, responsive, and private assistive technologies. In this work, we present SilentWear, a fully wearable, textile-based neck interface for EMG signal acquisition and processing. Powered by BioGAP-Ultra, the system enables end-to-end data acquisition from 14 differential channels and on-device speech recognition. SilentWear is coupled with SpeechNet, a lightweight 15k-parameter CNN architecture specifically tailored for EMG-based speech decoding, achieving an average cross-validated accuracy of 84.8$\pm$4.6% and 77.5$\pm$6.6% for vocalized and silent speech, respectively, over eight representative human-machine interaction commands collected over multiple days. We evaluate robustness to repositioning induced by multi-day use. In an inter-session setting, the system achieves average accuracies of 71.1$\pm$8.3% and 59.3\pm2.2% for vocalized and silent speech, respectively. To mitigate performance degradation due to repositioning, we propose an incremental fine-tuning strategy, demonstrating more than 10% accuracy recovery with less than 10 minutes of additional user data. Finally, we demonstrate end-to-end real-time on-device speech recognition on a commercial multi-core microcontroller unit (MCU), achieving an energy consumption of 63.9$\mu$J per inference with a latency of 2.47 ms. With a total power consumption of 20.5mW for acquisition, inference, and wireless transmission of results, SilentWear enables continuous operation for more than 27 hours.

Radio-based simultaneous localization and mapping (SLAM) has the potential to provide precise user equipment (UE) localization and environmental sensing capabilities by exploiting radio signals. Most existing approaches leverage line-of-sight (LoS) and single-bounce non-line-of-sight (NLoS) paths solely, while higher-order NLoS paths are treated as disturbance. In this paper, we investigate the benefits of leveraging double-bounce NLoS paths for solving the bistatic snapshot radio SLAM problem. We derive the Cramer-Rao bound (CRB) for joint estimation of the UE state and landmark positions when double-bounce NLoS paths are present. In addition, we propose an algorithm to identify double-bounce NLoS paths and leverage them into joint UE and landmarks estimation. The derived bounds are validated through simulated data, and the proposed algorithms are evaluated using experimental millimeter wave (mmWave) measurements harnessing beamformed 5G cellular reference signals. The numerical and experimental results demonstrate that the double-bounce NLoS paths which share at least one incidence point (IP) with the single-bounce NLoS paths improve the estimation accuracy of the UE state and existing IPs of single-bounce NLoS paths. Importantly, exploiting double-bounce NLoS paths enhances environmental mapping capabilities by revealing landmarks that are unobservable with single-bounce NLoS paths alone.

The DIarization and Speech Processing for LAnguage understanding in Conversational Environments - Medical (DISPLACE-M) challenge introduces a conversational AI benchmark for understanding goal-oriented, real-world medical dialogues. The challenge addresses multi-speaker interactions between frontline health workers and care seekers, characterized by spontaneous, noisy and overlapping speech. As part of the challenge, medical conversational dataset comprising 40 hours of development and 15 hours of blind evaluation recordings was released. We provided baseline systems across 4 tasks - speaker diarization, automatic speech recognition, topic identification and dialogue summarization - to enable consistent benchmarking. System performance is evaluated using diarization error rate (DER), time-constrained minimum-permutation word error rate (tcpWER) and ROUGE-L. This paper describes the Phase-I evaluation - data, tasks and baseline systems - along with the summary of the evaluation results.