Papers

Decoding and interpreting naturalistic speech from the brain increasingly relies on alignment to pretrained speech and language representation spaces. However, current CLIP-style brain-speech alignment ground neural activity to a single anchor modality-audio or text-despite the brain's inherently multimodal speech processing. This induces a trade-off: audio anchoring preserves temporal structure but weakens linguistic separability, while text anchoring captures semantics yet discards acoustic detail. We propose BAT-CLIP, the first CLIP-style trimodal alignment framework for iEEG that jointly aligns neural embeddings to both pretrained audio and text anchors in a shared, frozen audio-text manifold. On the naturalistic Podcast benchmark, BAT-CLIP yields more robust representations than bimodal CLIP baselines. We also highlight the importance of using self-supervised foundation models for CLIP training.

Neural receivers outperform conventional 5G NR processing chains, but their compute and memory demands hinder real-time deployment. For a standard-compliant multi-user MIMO neural receiver, the 4-bit number format, not merely the bit width, determines whether compression preserves the gain over classical receivers. We apply weight and activation quantization-aware training (QAT) and, separately, 50% magnitude pruning, comparing INT8/INT4 and FP8 (E4M3)/FP4 (E2M1) weights with INT8 post-ReLU activations. Trained on 3GPP UMi channels and evaluated on TDL-B and TDL-C, 8-bit weight-activation models remain within 0.05 dB of FP32 at 10% and 1% block error rate (BLER). At 4 bits, uniform INT4 loses 3.3-3.7 dB and falls below LS-LMMSE, whereas FP4 more than halves this loss (1.3-1.4 dB) and still outperforms it by about 0.5 dB, even after pruning. FP4's denser near-zero grid matches the trained weight distribution, and FP4 avoids the residual-path over-pruning seen with INT4. An analytic cost model projects 66x fewer bit-operations and 8.8x less weight storage for pruned 4-bit-weight inference.

This paper presents a data-driven framework for joint observer design and sparse sensor scheduling for unknown nonlinear networked systems. The nonlinear dynamics are approximated online as a piecewise sequence of locally linearized discrete-time systems, resulting in a switched linear representation recursively identified through Subspace State-Space System Identification (4SID). To ensure consistency across regime transitions, an Orthogonal Procrustes alignment is introduced to promote coordinate consistency across consecutive regime transitions and mitigate artificial discontinuities caused by arbitrary state-space coordinate changes. Based on the identified local realizations, a dual-rate predictor-corrector observer is designed. The observer gain is computed through a convex optimization problem that jointly addresses estimation accuracy, sensor sparsity, and stability requirements. In particular, an L_{2,1}-norm regularization term promotes column sparsity in the gain matrix, enabling the automatic selection of informative measurement channels, while a spectral-norm constraint guarantees Schur stability of the estimation error dynamics. The proposed framework is first validated on a traffic network simulated in Aimsun Next. Results obtained on an 18-link network show that the observer accurately reconstructs macroscopic traffic states while significantly reducing the number of active sensors required for real-time estimation.

Speech-breath segmentation is a fundamental preprocessing step in respiratory audio analysis, enabling applications such as respiratory acoustic biomarker extraction, lung function prediction and disease monitoring. Existing approaches, including threshold methods, Fourier Transform-based techniques, and unsupervised and pretrained voice activity detection (VAD) models, primarily focus on speech detection and often classify breathing events as non-speech or silence, limiting their applicability for precise breath detection. To address this limitation, we propose BreathGRU, a semi-supervised Bidirectional Gated Recurrent Unit (BiGRU) framework specifically designed for speech-breath segmentation. The proposed framework combines frame-level acoustic feature extraction with bidirectional recurrent modelling, pseudo-label refinement and duration-constrained Segmental Viterbi decoding to produce speech and breath segmentation. BreathGRU was evaluated against the existing approaches, using manually annotated recordings. Performance was assessed using event-based, time-based, overlap-based, duration-based and boundary-based segmentation metrics. Experiment results demonstrated that BreathGRU achieved the highest breath event recall (0.83), the lowest onset-localisation error (0.14s) and the highest Mean Match Intersection over Union (0.81), with competitive overall segmentation performance compared to large pretrained VAD models like Silero. Qualitative evaluation on manually annotated recordings further showed close agreement between BreathGRU and manual annotation, with better breath detection compared to Silero. These findings demonstrate that explicit breath event modelling provides advantages over general-purpose VAD models and establish BreathGRU as an effective speech-breath segmentation framework which can be applied for respiratory audio analysis and pulmonary healthcare applications.

It is necessary to characterize chemical reaction networks (CRNs) accurately to facilitate engineering of both synthetic and naturally occurring biological systems. Estimation of multiple unknown parameters and states in CRNs requires first, an identifiability analysis, and then adoption of a suitable estimation technique. In this paper, we took an example of a reduced order gene expression system, performed parameter sensitivity analysis for multiple unknown parameters, and used a modified Rao-Blackwellised particle filter (RBPF) to estimate parameters and states. The proposed framework estimates parameters using a particle filter and states using an extended Kalman filter (EKF) with process noise covariance updated recursively based on chemical Langevin equation (CLE). We compared the accuracy of parameter estimation, error in state estimation, and whiteness of the innovation sequence for the proposed filter with fixed choices of noise covariance. We found that the RBPF with updated noise covariance finds a balance between these three criteria, demonstrating its suitability for joint state and parameter estimation for stochastic CRNs.

Renewable energy power plants are increasingly expected to provide ancillary services to the grid. Yet, the variability of sources such as solar photovoltaics requires additional operational flexibility to deliver these services. In this context, hybrid energy storage systems (HESSs) offer an appropriate solution, because different technologies with complementary characteristics can be applied and leveraged to share the power demand and improve the overall performance. Yet, a proper power sharing requires a detailed model of the storage elements. This aspect has barely been studied in the literature, where most models consider constant efficiencies. Moreover, most formulations are based on optimisation problems that are computationally too demanding to be solved in real time. In this paper, a controller is proposed to minimise the losses of a HESS considering detailed power-dependent efficiency curves of each storage technology. The power sharing method is based on the analytical verification of the Karush-Kuhn-Tucker (KKT) conditions, which makes it computationally efficient and suitable for real-world deployment. The proposed controller is applied to a test case consisting of a 10 MW PV power plant with a HESS based on a lithium-ion battery and a redox-flow battery, each with 2.5 MW power and 5 MWh capacity. The main contributions are verified via numerical simulations performed in MATLAB/Simulink, whereas a real-time implementation deployed in OPAL-RT demonstrates the viability of the algorithm for real-time applications.

Pinching-antenna systems (PASS) create reconfigurable wireless channels by moving dielectric antennas along a waveguide. Mechanical position errors translate into phase errors that degrade performance, yet almost all prior PASS studies assume ideal placement. This paper models this position error, derives the resulting phase error, and analyzes its impact on time-division multiple access (TDMA) and non-orthogonal multiple access (NOMA), obtaining their ergodic-rate bounds and closed-form outage probabilities. Analyses and simulations reveal that single-antenna TDMA is immune to position error, whereas multi-antenna TDMA loses array gain and NOMA suffers an interference floor that cannot be overcome by increasing transmit power.

This paper studies constrained optimization problems through the lens of feedback control. Building on the interpretation of Lagrange multipliers as feedback controllers, we propose the saddle-point PID (SPPID) dynamics: a unified proportional–integral–derivative (PID) framework for continuous-time saddle-point dynamics. The proposed dynamics employ PID control for equality constraints and anti-windup PI control for inequality constraints. We show that SPPID is equivalent to a preconditioned primal–dual gradient flow of the augmented Lagrangian. This equivalence reveals the distinct role of each feedback component: integral action enforces constraint satisfaction, proportional action induces the augmented Lagrangian structure, and derivative action modifies the geometry of the primal dynamics via a state-dependent Riemannian metric. For convex problems, we establish convergence to the KKT set together with conditions for global asymptotic stability. For equality-constrained problems, we establish local exponential convergence for nonlinear equality constraints under standard assumptions, and show that projected gradient flow emerges as the infinite derivative-gain limit. For strongly convex problems with affine equality or inequality constraints, we establish global exponential convergence by leveraging contraction theory. Finally, we provide various numerical examples to illustrate the utility of SPPID.

Quantum Machine Learning is a novel field of research aimed at devising machine learning approaches exploiting principles of quantum mechanics, such as superposition, entanglement and interference. In this context, we present a scalable hybrid Quantum Diffusion Model, and evaluate its use for medical image analysis. Specifically, our method is based on a Discrete-Time Quantum Walk algorithm, executed on a real quantum device, to model the forward dynamics of the diffusion model. For the backward step of the diffusion model, we devise and evaluate a classical learning model, which is used to reversely denoise the data. In contrast with other existing attempts at applying quantum machine learning for image analysis tasks, severely limited by the size of existing quantum devices, our method allows to process real-world large size medical data. In particular, we present results on grayscale and RGB images, as well as 3D volumes of moderate sizes. We benchmark our results by reproducing an alternative classical counterpart model, based on diffusion models on discrete state spaces. By doing so, we compare the generation capabilities of both models in terms of three distinct state-of-the-art metrics in the field of image generation, showing the competitive, promising results of our approach.

Conventional tractometry averages white matter microstructure into a single profile, masking intra-bundle spatial heterogeneity and diluting localized alterations. We propose an automatic sub-bundle tractometry framework using Fréchet-based hierarchical clustering to decompose bundles into geometrically coherent sub-bundles, resolving complex fanning configurations without empirical thresholds. Multi-compartment metrics (fractional anisotropy, FA; isotropic free-water fraction, IFW) are projected onto these clusters. We evaluated this framework in 140 participants across early-life (ELD) and late-life (LLD) depression cohorts, alongside age-matched healthy controls. We assessed performance based on microstructural homogeneity, sensitivity to age, and clinical group differences. Proposed sub-bundle metrics were significantly more homogeneous than classical full bundle profiles (p < 0.001). Our approach showed enhanced sensitivity, detecting more age-correlated bundles in elderly controls (11 vs. 5) and more clinical group differences in LLD (FA: 6 vs. 4; IFW: 23 vs. 17) with larger effect sizes. In ELD, it improved the detection of subtle FA differences between remitted and resistant individuals (2.9% vs. 1.8%). Gains were particularly pronounced in LLD, reflecting the framework's ability to resolve fanning tracts susceptible to brain aging. By preserving spatial heterogeneity, this automatic decomposition provides a robust foundation for large-scale clinical tractometry.

We propose a self-supervised approach for learning room impulse response (RIR) representations from single-channel noisy-reverberant speech. It consists of first training on reverberant data, then on noisy-reverberant data, and finally with a teacher-student approach, where the student learns to replicate the teacher's embeddings when given a noisy version of the reverberant input. We assess their representational capabilities by estimating acoustic room parameters from them. Conditioning a discriminative speech enhancement model on the derived embeddings yields consistent gains across all evaluated metrics, including downstream word error rate, for both reverberant and noisy-reverberant speech.

Precise, high-speed control remains challenging for robots with complex actuation dynamics. Learning directly on hardware is further constrained by the cost of real-world interaction. We present an online model-based reinforcement learning framework that learns a probabilistic dynamics ensemble model from scratch for sampling-based model predictive control. A precision-gated contouring objective conditions the progress reward on path accuracy, prioritizing precision over speed. In a data-driven excavator simulator, the framework achieves higher sample efficiency than the evaluated model-based reinforcement learning baselines. We validate the framework by learning directly on an 11.5-ton Menzi Muck M445 hydraulic excavator, without demonstrations or simulation pretraining. After 20 minutes of interaction, the controller reaches tracking accuracy comparable to prior learned controllers trained on 100-150 minutes of data. After 40 minutes, it sustains sub-centimeter mean path error at high operating speeds.

Sound matching can be formulated as optimizing synthesizer parameters against an audio-domain objective. However, objectives derived from generic audio representations are often difficult to optimize, while direct search requires rendering every candidate. We introduce Synth-JEPA, which learns mutually predictive audio and parameter representations from paired synthesizer data. At inference, candidate parameters are scored directly in this learned space, yielding a renderer-free objective whose audio geometry is shaped by parameter correspondences rather than generic audio similarity. We evaluate Synth-JEPA on Surge XT using held-out synthesizer sounds and out-of-domain NSynth and FSD50K targets, against inverse models, direct search, and learned proxy objectives. Synth-JEPA outperforms all baselines in-domain and remains competitive out-of-domain. Its matching quality continues to improve with additional test-time search, allowing compute to be traded for match quality. In pairwise listening tests, listeners preferred Synth-JEPA in 85% of trials overall. Together, these results show that an audio representation with a parameter-induced geometry allows synthesizer sound matching to be approached as an effective renderer-free search problem.

Virtual power plants (VPPs) are an effective solution for increasing the penetration of renewable energy sources (RES) in power systems. Moreover, VPPs can provide ancillary services if generators and loads are closely connected and properly coordinated. This aspect has been addressed in several studies, yet they mostly focus on the dispatch optimisation level, leaving real-time operation aspects unresolved. To close this gap, a central controller for VPPs based on a rolling-horizon optimisation is proposed in this work. Its main objective is to maximise revenues while delivering frequency restoration services. A day-ahead optimisation firstly defines the power setpoint and the upward and downward reserves of the VPP. Then, the rolling-horizon optimiser, executed every minute, distributes the power and reserve references between the VPP units to achieve their close tracking while taking into account resource availability (wind speed, irradiance, etc.), operational constraints and battery degradation. The effectiveness of the controller is tested using a two-area interconnected power system under different RES generation and demand profiles. The simulations are performed in MATLAB/Simulink and include detailed dynamic models of the VPP elements, demonstrating the applicability of the algorithm to real systems. The day-ahead and the rolling-horizon optimisation problems are modelled using YALMIP and solved using Gurobi. The obtained results demonstrate the VPP supporting grid frequency restoration by optimally allocating its resources, even under the presence of forecast uncertainty.

Efficient acquisition of correlated continuous-time signals is critical in applications such as distributed sensor networks and array processing. However, existing correlation models often fail to capture the shared structure of physical signals measured in close proximity. In this paper, we model each signal in an ensemble as the sum of a hidden common lowpass component and a signal-specific innovation highpass component with disjoint spectral supports, where the common bandwidth is unknown. We first derive theoretical identifiability conditions guaranteeing unique decomposition and reconstruction from subsampled observations. Further, we propose a practical recovery algorithm and a joint reconstruction framework that leverages parametric structured dictionaries parameterized by the unknown common bandwidth. Numerical experiments on an ensemble of four signals demonstrate exact reconstruction with a 25% aggregate sampling rate reduction under identifiability conditions, and robust recovery with up to a 74% rate reduction when all channels are sampled below the Nyquist rate, significantly reducing data acquisition and hardware overhead.

Recent audio deepfake detectors separate bona fide speech from synthetic speech, yet it remains unclear which stage of a text-to-speech system supplies the detection evidence. We address this with controlled resynthesis and detector adaptation in an F5-TTS-BigVGAN pipeline. Since vocoder reconstruction of a real mel can itself be separable from the source utterance, we fix the vocoder and trace the larger change in detector separation to acoustic generation. Adversarially fine-tuning the acoustic model, with no detector in its objective, raises EER against fixed detectors at comparable quality. However, adapting a detector only on the tuned model's VCTK outputs lowers its LibriSpeech EER from 19.42% to 7.46% and improves detection of unseen base F5-TTS outputs. These results suggest that acoustic-model updates can reduce the detection evidence available to fixed detectors, while detector adaptation keeps the updated outputs detectable in this pipeline.

Speech content representations are central to voice conversion, speech-to-speech translation, and multimodal language models, yet they are rarely compared under a common generative framework that directly measures what each representation contains. We address this by training a generative model conditioned solely on each representation and evaluating the generated audio along the content, speaker identity, and prosody axes. Across SSL features, supervised tokens, posteriorgrams, and neural audio codecs, we find two distinct regimes: representations that nearly reconstruct the original audio, and representations that effectively disentangle speaker identity. These results show that disentanglement depends not on supervision alone, but on the interaction between the training objective and the representation's information capacity: supervised representations only disentangle speaker identity when their capacity is sufficiently constrained.

Lipschitz continuity is a fundamental principle in the design of certifiably robust deep neural networks (DNNs), wherein adjusting the Lipschitz constant, which quantifies network robustness, is of central theoretical importance. A standard approach to enforcing Lipschitz continuity requires each layer of a DNN to be Lipschitz continuous, thereby guaranteeing overall Lipschitz continuity. However, this layer-wise approach typically imposes overly conservative restrictions by producing a loose estimate of the overall Lipschitz constant, which limits the expressive capacity of the DNN and degrades empirical performance at a prescribed level of robustness. To overcome this loose estimation, the recently proposed LipKernel transfers information across layers to yield a much tighter overall Lipschitz bound than conventional layer-wise construction. In this paper, we extend this concept to cascaded state-space models (SSMs) to construct Lipschitz-continuous DNNs capable of modeling longer-term dependencies. The proposed architecture, named LipSSM, is theoretically justified and empirically evaluated.

Meta-adaptive filtering (Meta-AF) provides a data-driven alternative to hand-crafted adaptive filter updates by employing a learned optimizer throughout online adaptation. However, when Meta-AF is used for active noise control (ANC), time-varying acoustic paths remain a major challenge. In particular, the physics-informed optimizer features are constructed using a secondary path estimate and become mismatched when the physical path changes, leading to inaccurate filter updates and degraded noise reduction. To address this problem, this paper proposes a Coupled Meta-Adaptive Filtering Active Noise Control (CoMeta-AF-ANC) method, which applies meta-learning to jointly learn the control filter adaptation and acoustic path tracking within a closed-loop framework. The proposed Meta-Gated Joint Path Identifier (MG-JPI) simultaneously tracks the primary and secondary paths from the available ANC signals without auxiliary noise, while the updated secondary path estimate is fed back to reconstruct the Meta-AF controller features. A delayless dual-rate realization performs learned adaptation at the frame rate while generating the control signal at the sampling rate in the time domain. Evaluation using measured headrest acoustic paths shows that CoMeta-AF-ANC outperforms representative ANC algorithms in tracking time-varying acoustic paths, while maintaining higher stability across unseen head movement scenarios. It also generalizes well to real-world noises not encountered during training.

Simplicial complexes (SCs) extend graphs to represent higher-order interactions, but couple their simplex levels through the inclusion property. While static SC inference is an emerging research direction, tracking time-varying SCs remains largely unexplored. A central challenge in tracking SCs is to account for the inclusion property. To this end, we build a nonlinear state-space model tailored to SCs. For prediction, we introduce a closure-aware Markov generation model whose conditional mean preserves simplicial inclusion and whose covariance captures edge-triangle dependencies. For correction, we encode inclusion constraints as nonlinear pseudo-measurements, allowing the constraints to inform both the state and covariance updates. We investigate progressively richer treatments of these measurements through a standard extended Kalman filter, an iterated extended Kalman filter, and a Laplace approximation. Experiments then demonstrate the benefits of exploiting the proposed dynamics and constraint information.

Accurate and efficient pronunciation transcription is essential for preparing text-to-speech training data at scale. Existing approaches have different limitations: grapheme-to-pronunciation (G2P) and speech-to-pronunciation (S2P) methods each capture only partial information, using only text or only speech, while speech-and-text-to-pronunciation (ST2P) methods use both but require costly pronunciation-annotated data. To address this problem, we propose a training-free ST2P pipeline that integrates both lexical and acoustic information at inference time. Lexical resources and G2P tools generate text-constrained candidates, and a left-to-right greedy search selects the best one using whole-sequence negative log-likelihoods from frozen pretrained S2P models. On three Japanese corpora, our method reduces Character Error Rate (CER) from 0.60–1.40% (text-only baseline) to 0.04–0.17% with reference transcripts, and 0.64–1.58% with ASR transcripts. It outperforms all baselines, including a trained ST2P model and commercial multimodal LLMs. Our greedy search method is 3–3.5$\times$ faster than beam search at similar CER, and the cascade is 2$\times$ faster than direct decoding ensuring the efficiency and accuracy. In Spanish, French, and preliminary English, it also surpasses four open multimodal LLMs and the best traditional methods.

Music source separation (MSS) methods aim to extract stems from music mixtures, which is important, for example, in karaoke, music remixing, and pedagogical applications. While earlier research on MSS systems has been dominated by models targeting narrow sets of general stems, there have recently been attempts to support broader source definitions. One of these methods uses audio queries to provide direct and descriptive control over the desired separation targets based on the sound itself. However, query-based separation remains cumbersome due to the need to provide audio examples with features matching the sources contained within mixtures. This paper proposes Music Source Separation via Stem Discovery (MuS3D), a query-based source separation framework that iteratively discovers active sources from the mixture. On correctly detected sources, our model matches manually queried baselines and surpasses state-of-the-art text-based models. Subjective evaluation indicates encoding artifacts as the limiting factor in current generative separation. The findings suggest that audio-based query representations offer an effective and automatable interface for source separation.

Semantic communication (SemCom) and integrated sensing and communication (ISAC) are promising technologies for future 6G wireless networks. Existing studies have applied semantic technology to either the communication module or the sensing module of ISAC. In this work, we propose SemISAC, which performs both SemCom and semantic sensing within a single dual-function waveform. SemISAC uses a joint semantic encoder that extracts task-specific information for both communication and sensing. We evaluate SemISAC in a vehicular scenario in which vehicles share pixel-wise segmentation of the road environment and, through sensing, classify surrounding objects and estimate their ranges. On the transmitter side, a deep learning encoder converts the input road-scene image into semantic symbols and places them on the data cells of an OFDM grid, while the remaining cells serve as pilots for channel state information estimation and sensing. The pilot configuration is adaptively optimized based on the channel conditions to balance communication and sensing requirements. At the receiver, a deep learning model reconstructs the segmentation from the received waveform, while the transmitting vehicle captures the reflected waveforms from surrounding objects and uses task-specific deep learning decoders for target recognition and range estimation. Simulation results show that SemISAC achieves a segmentation accuracy close to that of the dedicated SemCom module while outperforming both conventional and semantic baselines in target recognition and range estimation.

A series elastic actuator has a single torque port and pays for it twice: the geared motor must swing its own reflected inertia through the spring, so the amplitude it delivers collapses as $\omega^{-2}$ in the command frequency $\omega$ once it saturates, while commands below the transmission's breakaway friction never arrive at all. This letter opens a second torque port on the load side, placing a frameless direct-drive micro motor in parallel with a fixed-stiffness spring – a parallel-integrated SEA, or Pi-SEA, whose delivered torque is read from spring deflection and micro current without a sensor – and dividing the commanded torque between the two channels by time scale rather than by filter design. The micro torque loop is the fast subsystem, which makes the closed loop singularly perturbed and turns the separation the channels need into a bound to check rather than a crossover to tune; a leaky mid-ranging integrator returns the steady load to the spring; and the amplitude ceiling, read backwards, becomes a closed-form sizing rule that matches spring, geared motor and micro motor to the amplitudes and frequencies an application asks for. Against SEAs, the Pi-SEA widens the tracked band at small amplitudes and lowers the residual the joint imposes on its environment, each by an order of magnitude.

Contingency analysis using the AC power flow (AC-PF) model is a critical tool for accurate grid security assessment, but its computational burden increases with the number of operating scenarios and outage configurations to evaluate. Recent ML-based approaches typically require outage-specific training data, leading to offline training costs that scale with the number of contingencies. This work proposes a framework that reuses a single ML model trained solely on basecase AC-PF data to estimate post-contingency operating states under arbitrary single-line outages. The proposed approach formulates post-contingency state prediction as a fixed-point iteration. If the ML model is a deep neural network (DNN), we derive sufficient conditions that guarantee convergence and develop semidefinite programming (SDP) formulations to certify these conditions for a given DNN. Numerical tests on the IEEE 118-bus system demonstrate that the proposed SDP formulations are tight, that the certified conditions hold for all tested contingencies, and that the resulting method produces accurate post-contingency state estimates within only a few iterations.

High photovoltaic (PV) penetration planning requires repeated nonlinear PV and power-flow evaluations under uncertain irradiance and demand, particularly when two competing objectives, penetration and total electrical loss, are optimized simultaneously subject to reliability constraints. This paper develops a decision-focused, surrogate-assisted stochastic multiobjective framework in which a calibrated PV model coupled with a distribution-network simulator provides the high-fidelity reference, while matched data-only and physics-informed surrogates accelerate optimization and are evaluated empirically for decision reliability across the loss-penetration tradeoff. Although the two surrogates exhibit nearly identical held-out errors, optimizer-realistic screening yields markedly different operational validity rates of 44.4% and 93.4%, respectively, motivating selection of the physics-informed model on decision performance rather than test accuracy. Surrogate inference cuts the cost of a 60-scenario design evaluation from 286.0 to 0.46 ms, an approximately 620 time speedup. A decision-gated enrichment strategy further targets Pareto-critical, constraint boundary, and surrogate-disagreement regions, reducing prediction error by 39.1%. Independent 500 scenario certification verifies all 80 screened candidates and identifies 46 non-dominated designs; the selected compromise attains 70.74% expected penetration with 297.9 kW expected total electrical loss. Cross-feeder transfer eliminates baseline voltage violations on the 85- and 69-bus systems, but produces overvoltage and a 62.7% increase in feeder-network loss on the 33-bus system, demonstrating that certified penetration is topology-specific and demands decision-level auditing, independent certification, and feeder-specific validation.

In this paper, we propose I-Parakeet, an integer-only implementation of NVIDIA's Parakeet-CTC (0.6B parameters) that runs on a smartphone NPU without any floating-point operator or CPU fallback. Modern Conformer ASR models are hard to deploy on edge devices because of their size, and quantized models still fall back to floating point for numerically sensitive operations. This prevents them from fully exploiting integer accelerators such as mobile NPUs. To achieve this, our contributions are threefold. First, we derive an integer formulation of the relative-positional self-attention at the core of the Conformer. We fuse its two score branches with different quantization scales and the relative shift into integer-only operations. Second, we introduce a minimax-optimized Swish approximation that minimizes the maximum error of the Swish output. Third, a layer-wise range analysis of activations yields two targeted remedies: an INT16 grid for the BatchNorm output and percentile calibration for the heavy-tailed pre-encoder activations. I-Parakeet achieves 4.97% WER on LibriSpeech test-other, running on a Qualcomm NPU at a real-time factor of 0.048, 7.5x faster than a CPU baseline.

Perceived speech decoding based on non-invasive brain-computer interface (BCI) signals has been extensively studied in recent years. Research in this field primarily faces two challenges: extracting neural representations with rich spatiotemporal information and achieving cross-subject generalization. Although separate studies have proposed methods to cope with these issues, a unified approach that simultaneously tackles both challenges remains lacking. To fill this gap, we propose the Subject-Invariant Cross-Modal Perceived Speech Decoding (SICMD) method, which integrates functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG). We conduct comprehensive analyses of the fusion method, fusion position, encoder architecture, and model inputs. Our results demonstrate that the proposed method improves Top-1, Top-10, and Rankacc by more than 10.6%, 10.1%, and 1.7%, respectively, compared to baseline methods in cross-subject perceived speech decoding tasks, while reducing training costs by 88.8% and 60.5% compared to multi-subject and intra-subject decoding settings. Further visualization experiments also confirm the effectiveness of our approach.

Reliable absolute pose estimation in urban environments is undermined by outlier measurements and incorrect temporal associations that can persist in tightly coupled estimators. Global Navigation Satellite System (GNSS) observations provide globally referenced measurements but are prone to multipath effects. Visual-inertial sensing supplies local motion constraints, but false visual associations can corrupt the estimator. We present SeA-RVINS, a fixed-lag factor-graph Real-Time Kinematic (RTK) visual-inertial system for robust urban pose estimation. A semantic-aware learned stereo frontend rejects unreliable tracks before persistent landmarks enter the graph. For double-differenced GNSS measurements, SeA-RVINS applies Dynamic Covariance Scaling through configurable batch, scalar, and latent-pivot robust formulations while retaining the shared-pivot correlation structure. We propose a hybrid ambiguity-continuation strategy that shares one ambiguity state over short arcs with verified continuity and softly links successive arcs through random-walk factors. On an approximately 20-km route from the public TEX-CUP dataset, including about 50% deep-urban driving, the latent-pivot configuration achieves 100% availability and a 1.6-m maximum horizontal error, with 96.16% and 99.90% of epochs below 1.0 and 1.5 m, respectively. The implementation is released as open-source software

This paper proposes a movable-element (ME) enabled simultaneously transmitting and reflecting surface (ME-STARS)-assisted non-orthogonal multiple access (NOMA) communication system, where the positions of STARS elements can be flexibly reconfigured to introduce additional spatial degrees-of-freedom. Our objective is to maximize the sum achievable rate by jointly optimizing the active beamforming at the base station (BS) as well as the passive beamforming and MEs positions at the STARS. To tackle the resultant highly-coupled non-convex optimization problem, an alternating optimization-based algorithm is proposed, in which the original problem is decomposed into three subproblems. For the BS and ME-STARS beamforming subproblems, the successive convex approximation method is employed. For the MEs positions optimization, penalty terms are incorporated into the objective function to handle the constraints, thereby transforming the original problem into an unconstrained formulation. The gradient descent algorithm is subsequently applied to iteratively approach the sub-optimal solution. Numerical results demonstrate that: i) ME-STARS-assisted NOMA communications achieve superior performance compared to the NOMA communications assisted by STARS with fixed-position elements; and ii) the performance gain brought by the MEs is significant even with restricted moving region size.