Papers

Safe robot navigation in crowded spaces requires planning that accounts for where people will be, not only where they are now. Model Predictive Path Integral (MPPI) control is an effective sampling-based planner, but many implementations encode humans as static point obstacles at their current positions, underestimating risk in dynamic scenes. We propose Predictive Gaussian Interaction Fields (PGIF), a spatiotemporal cost formulation that propagates pedestrian predictions forward over the full planning horizon and encodes them as anisotropic Gaussian repulsive fields aligned with each pedestrian's direction of motion. The forward spread of each field grows with the pedestrian's speed, creating a motion cone danger zone that penalises robot trajectories entering the pedestrian's path of travel more strongly than those approaching from behind. The formulation is closed-form and fully parallelisable across rollouts, adding no measurable computational overhead. Evaluated over 300 randomised crowd scenarios at three density levels, PGIF-MPPI achieves a 0% collision rate at every density level, compared with up to 82% for vanilla MPPI, while maintaining real-time planning performance.

Reliable identification of railway wheel defects is important for safety and maintenance. This study develops a machine-learning-based diagnostic framework for multi-class defect identification using passive air-coupled ultrasonic acoustic emission signals. Data were collected from eleven full-scale railway wheelsets representing nine health states. Time- and frequency-domain features were evaluated using Kruskal-Wallis statistical testing and mutual-information analysis to identify the most discriminative indicators. A Random Forest classifier was then trained using the selected features with stratified 5-fold cross-validation. The model achieved a balanced accuracy of approximately 0.66 and a Macro-F1 score of 0.65 across the nine classes. Decay rate, kurtosis, skewness, and envelope low-frequency power emerged as the most influential features, while a compact subset of features retained most of the classification performance. The results demonstrate the feasibility of combining passive ultrasonic sensing, statistical feature selection, and supervised machine learning for non-contact railway wheel defect classification and provide a foundation for future field-deployable inspection systems.

This paper presents a scalable lateral control framework for robust path tracking of single-body and articulated autonomous vehicles at high speeds. A clothoid-based controller is extended with three key adaptations: 1) an integrated tangential check and Frechet distance method for optimized lookahead and oscillation mitigation; 2) real-time trajectory segment classification for dynamic adjustment of the lookahead search range; and 3) a dual-adaptive, rate-controlled lookahead mechanism responsive to cross-track error and upcoming path geometry. To support different vehicle configurations, the framework also incorporates flexible tracking point selection and curvature-to-steering lookup tables, enabling control of points such as the tractor rear axle, hitch point, or trailer-related locations without changing the core control architecture. The controller is evaluated using high-fidelity TruckSim and Simulink simulations for both standalone and articulated vehicles across dual lane changes and winding-road scenarios. Results demonstrate stable high-speed path tracking, reduced oscillations, effective lookahead adaptation, low cross-track errors, and acceptable lateral acceleration across different vehicle configurations. The results support the use of a unified lateral control architecture that can scale from single-body to articulated autonomous vehicles.

Neural speech codecs face a fundamental tension in the language-model era: tokens that support high-fidelity reconstruction are not necessarily easy for autoregressive models to predict. Our controlled analysis of diverse codec and self-supervised speech representations shows that clearer phoneme structure before discrete code assignment is consistently associated with easier autoregressive token prediction. Yet phoneme structure alone is insufficient for high-fidelity reconstruction, which also requires reconstruction-relevant acoustic detail. Guided by this observation, we introduce ReLMCodec, a low-bitrate single-codebook speech codec built upon a preserve–control–refine principle: it preserves the linguistic organization of frozen self-supervised learning (SSL) features at the quantizer input, controls reconstruction-driven drift through Pre-quantization Anchor-Preserving Adaptation (PAPA), and refines the quantized latent space with a training-only WavLM-Large L24 teacher to reduce phoneme-level token fragmentation. Together, these components allow acoustic detail to support waveform reconstruction while keeping the resulting token sequence predictable for autoregressive models. At 650 and 800 bps, ReLMCodec moves the empirical single-stream predictability–reconstruction frontier in our evaluations, with gains that carry over to downstream text-to-speech (TTS) synthesis in both intelligibility and speaker similarity.

India's linguistic landscape spans over 700 languages and thousands of dialects, yet the vast majority of automatic speech recognition (ASR) systems support only a small fraction of this diversity. We present SraVaani-1.0, a multilingual ASR model covering 65 Indian languages and dialects, many of which currently have no publicly available or competing ASR system. SraVaani-1.0 is built on a FastConformer architecture and trained from scratch through a three stage the first stage, we perform self-supervised pretraining on 31,255 hours of unlabelled speech from the VAANI corpus using a contrastive learning objective. In the second stage, we introduce an audio-image representation alignment stage that leverages the paired images and speech available in the VAANI corpus. This multimodal alignment encourages the speech encoder to learn semantically richer representations by exploiting the relationship between visual context and spoken content, thereby improving downstream recognition, particularly for low resource the final stage, the aligned encoder is fine-tuned end-to-end using a Hybrid Token-and-Duration Transducer (TDT)-CTC decoder on 31,263 hours of labelled multilingual Indian speech compiled from 24 public datasets spanning 65 languages and dialects. We evaluate SraVaani-1.0 against three state-of-the-art multilingual ASR systems across eight benchmarks. SraVaani-1.0 achieves the lowest word error rate (WER) on a large number of language-dataset pairs while remaining competitive with the best-performing systems on high resource importantly, it is the only open-source evaluated model that provides transcription capability for multiple low-resource and tribal Indian languages, which are assessed exclusively on the VAANI benchmark.

The rapid growth of low-altitude aerial services and applications, driven by uncrewed aerial vehicles (UAVs), calls for a new class of digital infrastructure beyond conventional terrestrial networks. The low-altitude wireless network (LAWN) has been proposed as dynamically reconfigurable three-dimensional architectures that integrate aerial and ground nodes to provide connectivity, sensing, and control in open, safety-critical airspace. This tutorial presents a comprehensive treatment of LAWNs from the joint perspectives of artificial intelligence (AI) and signal processing. We first review the historical evolution and architectural foundations of LAWNs, introducing altitude-based layers and functional planes, and summarizing the regulatory and standardization landscape. Building on this system view, we then discuss signal processing fundamentals for LAWNs, including 3D channel and system models, performance metrics, waveform and receiver design, localization and tracking, and multi-functionality co-design. Next, we survey AI techniques for LAWNs, covering discriminative and generative models for perception, control, resource management, and security, as well as emerging paradigms such as foundation models, large language models, and digital twins for mission planning and closed-loop optimization. To illustrate AI-signal processing integration in practice, we provide a case study of an AI-driven multi-tier LAWN with hybrid satellite, high-altitude, and ground nodes. The tutorial concludes by outlining key research challenges in architecture design, signal processing-AI co-design, safety and security, experimentation, and standardization, and by highlighting opportunities for LAWNs to evolve into dependable, AI-native infrastructure for the intelligent skies.

Growing shares of variable renewable energy sources (VRES) increase forecast uncertainty and variability, requiring flexibility reserves that adapt to changing operating conditions and reflect the likelihood and severity of forecast deviations. Conventional fixed-rule or Gaussian approaches misrepresent asymmetric, heavy-tailed errors, leading to inefficient procurement or optimistic risk estimates. This paper presents a source-agnostic framework that constructs conditional error distributions for load, wind, and solar from historical deviations or probabilistic forecasts, combines them into a conditional net-load error distribution, and derives upward and downward reserves using coverage- and risk-based criteria. Nonparametric density estimation captures empirical error behavior without a parametric assumption, while a Conditional Value-at-Risk (CVaR)-based metric quantifies expected uncovered deviations. Experiments on a New York Independent System Operator (NYISO)-based synthetic dataset show that the framework meets target coverage with lower reserve volumes than static benchmarks and avoids the tail-risk distortion of Gaussian-based methods. Because reserves are built directly from the resource uncertainty distributions, the framework is transparent and interpretable, and it integrates seamlessly into existing production-cost models as deterministic reserve constraints, avoiding scenario-based stochastic optimization. It is implemented in the Electric Power Research Institute's (EPRI) DynADOR tool for operational reserve scheduling.

Recent low-light image enhancement (LLIE) methods have driven brightness and structural fidelity close to that of normally-exposed images, yet their outputs still exhibit systematic color shifts such as greenish skies, yellowish faces, and warm-tinted white objects. We attribute this to end-to-end LLIE training coupling brightness, structure, and color within a single network, leaving the color channels weakly supervised. We recast color restoration as an independent sub-problem and decouple it from brightness enhancement, realizing it as a general-purpose post-processing module built on retrieval-augmented generation (RAG). Rather than relying solely on parametric color priors learned during training, the module dynamically retrieves a reference image from an external high-quality color knowledge base and injects its color distribution into a color-restoration network to correct residual bias. The design has three components: (i) a dual-index FAISS retriever built on intermediate VGG19 features, capturing textural and structural similarity through global mean and variance statistics; (ii) GlobalSPHistAdaIN, which reduces the reference spatial-preserving color histogram to a global color vector and modulates network features via adaptive instance normalization, removing dependence on pixel-level correspondence; and (iii) a residual formulation that predicts a color correction over the front-end output. Across LOLv1, LOLv2-Real, and LOLv2-Synthetic, the module consistently improves color-specific metrics, and it remains effective when the front end is swapped among CPGA-Net++, LLFormer, FLIGHTNet, and IAT, confirming cross-front-end generality. Ablations show that a VGG19 dual index outperforms CLIP-based retrieval, indicating that color restoration depends on textural and structural similarity rather than high-level semantics.

Communication delay remains a central challenge in telerobotics, where it disrupts visuomotor coordination and reduces task precision. Motion scaling is an effective countermeasure to delay-induced overshoot, yet typical deployments rely on uniform gains that neglect individual and contextual variability. We propose a human-centered method that fits personalized delay-, direction-, and distance-specific scaling parameters for each participant. We conducted experiments with twenty participants who performed delayed reaching tasks in a virtual simulator. Scaling gains were computed to minimize mean overshoot in simulation in each combination of experimental conditions. Evaluation was done in simulation and on a telesurgical robot to evaluate assistance benefits. Performance was assessed across multiple delays, distances, and movement directions using overshoot, endpoint error, trajectory smoothness, economy of motion, and a composite error-time metric. Motion scaling consistently improved performance relative to unassisted trials, yielding up to 20-25% performance gains in key metrics. Effects were most pronounced at longer delays. Personalization demonstrated additional accuracy benefits for inward reaching at a short distance under moderate delay. The results highlight the potential of personalized scaling as a foundation for more adaptive frameworks that integrate contextual information to improve the safety and precision of teleoperated procedures.

Longitudinal MRI enables sensitive measurement of structural brain change for studying aging and neurodegenerative disease. Deformable image registration is a key tool for estimating such change by computing a dense deformation that captures geometric differences between longitudinal scans. However, MRI scanners introduce geometric distortions that vary across acquisition systems and protocols, such as gradient non-linearity (GNL) distortion. Existing registration methods estimate a single field that conflates biological and technical effects, potentially biasing downstream morphometric measurements if distortions remain (partially) uncorrected. We propose $\texttt{DisMorph}$, a registration framework trained entirely on synthetic data that explicitly decomposes longitudinal deformation into technical and anatomical transforms. It predicts two dense deformations, each encoding one effect. During training, a novel generative model synthesizes both effects separately to provide disentanglement supervision, while domain randomization promotes generalization across imaging protocols. We evaluate our method in three complementary settings. On simulated data with known ground truth, our method detects anatomical change more accurately and consistently than conventional registration. On real image pairs that differ only by GNL distortion, our method assigns most geometric change to the distortion field, demonstrating specificity in the absence of anatomical change. On longitudinal Alzheimer's disease (AD) pairs, our method detects anatomical change in AD-related brain structures while identifying residual distortion left after standard correction. By disentangling MRI-induced distortion from biological change in the longitudinal deformation, our method paves the way for more accurate longitudinal morphometry in clinical settings where maintaining acquisition consistency is challenging.

Interconnection queues, not electricity prices, now govern where data centers can be built, and the standard levelized-cost comparison answers a question no developer faces: it assumes a load profile, freezes the grid price while modeling the demand that moves it, and quotes busbar costs a facility cannot buy. This paper evaluates nine on-site supply technologies against a delivered grid whose price is endogenous to projected data-center demand, on a complete-site basis that retains standby charges, with measured GPU training load, delivered fuel prices, production-pathway carbon, and statutory 45V and 48E incentive mechanics. Nothing beats the wire: gas combined cycle produces at 47 USD/MWh but costs about 114 USD per megawatt-hour of complete site energy against a 92 USD grid; four-hour storage is physically capped near 18 percent of annual energy and, charged at the margin, dirtier than the grid; hydrogen from grid-priced power fails on cost and carbon together. An investment inversion converts these findings into capital terms: conversion-hardware learning buys nothing, because free hardware still exceeds the grid for every low-carbon arm, while global electrolyser deployment on sited sub-20 USD/MWh power brings PEM hydrogen power to about 2.2 times the grid at 300 billion USD and 1.9 times at 1 trillion USD (2.7 and 2.3 for the hydrogen engine), with a carbon reduction of roughly 85 percent (6.8-fold) against grid-power production. Grid parity is not purchasable at any budget. On-site supply is an access and depth product; most current investment targets the wrong term.

Cislunar spacecraft operate in nonlinear and unstable environments that make defensive maneuver planning difficult. We formulate cislunar spacecraft pursuit and evasion as a zero-sum differential game in the circular restricted three-body problem. Each spacecraft controls its thrust and reference orbit phase, enabling motion along periodic orbits and across quasi-periodic tori while remaining near the reference. We solve the game using a constrained discrete-time differential dynamic programming method that enforces hard input constraints. We propose a shared time regularization to synchronize both spacecraft while refining the discretization near close lunar passages. Numerical results on periodic and quasi-periodic orbits show that phase control improves maneuvering flexibility while limiting departure from the reference orbit. The discrete-time method also provides a large computational advantage over a continuous-time formulation. Comparisons between quasi-halo and quasi-near-rectilinear halo orbits show that close lunar passages create larger escape opportunities but also increase the sensitivity of the encounter. These results show that reference orbit geometry is an important part of defensive cislunar mission design.

Target sensing utilizing 5G NR reference signals has emerged as a prominent research direction in both academia and industry. However, non-ideal factors in practical deployments exert a significant detrimental impact on target sensing performance, manifesting as artifacts in the RV spectrum. These artifacts mask weak targets and cause severe false alarms. To address these challenges, this paper establishes a theoretical model of artifacts and constructs a data-physics-driven deep learning paradigm for artifact mitigation. First, the origin of artifacts and their characteristics are theoretically derived. These analyses demonstrate that the artifacts are associated with polyphase codes, e.g., Zadoff-Chu sequences, and reveal their characteristics, including periodic extensions in the range domain and spectral spreading in the velocity domain. Then, the physical priors of artifacts are formalized as temporal continuity and spatial consistency, informing the design of the training mechanism for the proposed network. Guided by these insights, we propose a PIAENet. At its core is a multi-frame selective-masked encoder-decoder module, explicitly designed to incorporate the above priors. Specifically, temporal continuity is implemented via a multi-frame mechanism to capture features across consecutive RV spectra. Meanwhile, spatial consistency is realized through a selective masking mechanism to enhance reconstruction of artifact-affected regions. Extensive validation is conducted using real-world measured data collected with commercial mmWave equipment. The polyphase-code-related characteristics of the artifacts are experimentally validated. Meanwhile, the experimental results demonstrate that the proposed PIAENet not only effectively reduces the false target count but also improves the detection probability from 79.58% to 98.88%.

Vehicular Ad hoc Networks (VANETs) increasingly rely on federated learning (FL) to enable collaborative intelligence without sharing raw sensory data. However, most existing vehicular FL frameworks assume that all vehicles train a single global model for a common task, which limits their applicability in practical vehicular environments where vehicles may perform heterogeneous learning tasks under non-independent and identically distributed (non-IID) data, intermittent connectivity, and high mobility. To address these challenges, this paper proposes an AutoEncoder-based Reliability-Optimized Hierarchical Multi-Task Federated Learning (AERO-HMTFL) framework for dynamic multi-hop clustered VANETs. The proposed framework introduces a tri-weighted clustering metric that jointly considers vehicular mobility, shared-model similarity, and task affinity to produce mobility-stable, semantically aligned clusters. Each vehicle employs a split-model architecture comprising a shared autoencoder-based representation module and multiple task-specific heads, with only the shared autoencoder parameters exchanged while the task heads remain local. To improve robustness, cluster heads perform reliability-aware aggregation based on historical validation performance and participation frequency, while the Evolved Packet Core (EPC) conducts global shared-autoencoder fusion across clusters. Extensive simulations demonstrate that, compared with the multi-task federated learning benchmarks, AERO-HMTFL achieves up to 13% higher sustained EPC-level accuracy, exhibits more stable learning dynamics, and reduces EPC-level packet transmissions by approximately 87-97%. Under short-range connectivity, it also requires approximately 13-29% fewer communication rounds to converge.

Root cause analysis (RCA) in microservice architectures needs to pinpoint the originating faulty service responsible for the cascading symptoms seen across hundreds or thousands of interdependent services. Graph-based random walk methods propagate anomaly evidence over the service dependency graph. However, existing anomaly-restart walks leave much of the localization signal unused: they restart from the raw anomaly vector, which is dominated by loud downstream victims rather than the quieter source. Through a controlled ablation, we first show that the "resilience damping" often applied to the transition matrix is mathematically equivalent to raising the restart probability; we therefore benchmark against a restart-tuned Personalized PageRank (PPR) rather than its default configuration. We then present Enhanced Iterative Random Walk with Restart (eIRWR), which (a) concentrates restart mass on the most suspicious nodes through power-law teleportation sharpening, (b) augments the transition matrix with self-loops and backward edges so that probability accumulates at cascade sources, and (c) refines its belief across an outer loop. On three large-scale topologies (12K-25K nodes) from the Alibaba Microservice Trace Dataset, eIRWR attains a Mean Reciprocal Rank (MRR) of 0.75 at moderate root-cause visibility, a 2.8 times improvement over the best aggregate-metric baseline and well above a restart-tuned PPR. At high visibility, it reaches MRR= 0.94, while running in under 25ms on graphs with 17,000 nodes, making it suitable for online deployment.

Motion deblurring has achieved remarkable progress with deep learning, yet pre-trained deblurring models often suffer from performance degradation in real-world scenarios due to the domain shift between training and testing distributions. To remedy this, we propose EvBS, an event-guided blur synthesis framework that generates diverse training pairs for calibrating pre-trained models to the target domain. While existing methods are constrained by the inherent entanglement between motion and visual content, our method leverages the high temporal resolution of event cameras to effectively decouple them. This enables us to utilize not only the intrinsic motion that is inherent to the given content but also extrinsic motion transferred from different sources within the target domain, thereby facilitating effective adaptation via fine-tuning. Specifically, EvBS comprises two complementary strategies: Intrinsic-Blur Synthesis, which blurs sharp contents with their own motion patterns, and Extrinsic-Blur Synthesis, which transfers motion from blurry patches to distinct sharp content. This approach generates a diverse set of training pairs that break the inherent constraints of naturally coupled motion and content, resulting in enhanced domain-adaptive deblurring performance. Extensive experiments on multiple benchmarks demonstrate that EvBS effectively enhances the robustness of existing deblurring models on unseen testing datasets.

This paper presents a parameter-dependent linear matrix inequality (LMI) framework for trajectory tracking of nonholonomic mobile robots subject to severe multiplicative wheel slip on variable-terrain surfaces. The sampled convex formulation, augmented with grid-to-continuum residual certification, simultaneously establishes semi-global differential input-to-state stability, a prescribed exponential decay rate, regional pole placement, and a gain-bounded feedback proxy for actuator-limited operation. A central contribution is an explicit upper bound on the additive disturbance induced by bounded multiplicative slip in the Kanayama error coordinates, bridging the physical slip mechanism and the convex synthesis paradigm. The auxiliary gain matrix and inverse storage metric are parameterized affinely in the reference velocities, while the storage metric inherits nonlinear dependence through pointwise matrix inversion. Stability is established via a cascade analysis combining variational contraction, forward invariance, slip-induced disturbance bounds, and dissipation-based trajectory reconstruction. Numerical validation compares three controllers across six reference trajectories, six disturbance classes, and a 60-second variable-terrain test featuring six severe slip patches with bidirectional slip ratios reaching +/-50%, replicated on two geometries. Supplementary studies address Gaussian sensor noise, compound stress-testing, and embedded-platform computational feasibility. Across 100 Monte-Carlo runs the proposed controller achieves complete trajectory containment within the certified envelope. On the variable-terrain scenario, peak tracking error is reduced by 12% against the fixed-gain LMI baseline and 49% against the manual baseline, with the constant-gain baseline infeasible at the prescribed decay rate.

This paper presents, for the first time in power systems literature to our knowledge, analytical tools to explain the training performance of machine learning surrogate models for power system dynamics. Power system simulations are increasingly challenged by stiff and multi-timescale dynamics arising from converter-interfaced resources and fast control loops. Machine learning surrogates emerge as promising tools to handle this complexity and accelerate dynamic simulations. However, their performance remains difficult to interpret, which limits their adoption. Building on the small-signal eigenvalue analysis in power systems, this paper uses the Neural Tangent Kernel (NTK) method. NTK delivers a modal interpretation of the learning performance, identifying error modes that decay rapidly versus others that converge slowly. This connection explains how physical stiffness and timescale separation in power system dynamic models appear as optimization stiffness during Neural Network (NN) training. Based on this analysis, we develop adaptive loss-weighting strategies to improve and explain why structure-aware neural architectures, such as ActNet, perform better than vanilla NNs. We assess the proposed approach on physics-informed machine learning surrogate models of \acp{SM} and power electronic converters. The methods introduced in this paper can deliver the necessary analytical tools to interpret and improve the performance of machine learning surrogates, paving the way for the systematic, physics-aware design of NN architectures and training strategies. By moving beyond trial-and-error development, these tools reveal training dynamics and failure modes, support more reliable design decisions, and strengthen confidence in machine-learning surrogates for engineering applications.

Wireless foundation models are emerging as a promising paradigm for AI-native physical-layer design. However, existing methods typically model channel state information (CSI) as image-like discrete tensors with generic token decoders that may struggle to capture complex high-frequency variations efficiently and often produce high-dimensional, size-dependent representations. In this paper, we propose WiFo-INR, an implicit neural representation (INR)-based wireless foundation model that represents CSI as a coordinate-conditioned neural function. A Transformer encoder maps partial or coarse CSI to fixed-dimensional modulation tokens that adapt a SIREN-based decoder, and a compression autoencoder enables quantized CSI feedback. It adopts a two-stage self-supervised pretraining scheme, where mixed masking and denoising improve channel reconstruction and compression-enhanced pretraining enables accurate CSI feedback at low compression ratios. Extensive experiments demonstrate that WiFo-INR learns efficient, compact, and CSI-size-independent implicit wireless representations. Compared with existing foundation models, WiFo-INR improves channel reconstruction and CSI feedback performance while substantially reducing inference latency. It also transfers efficiently to diverse wireless tasks with minimal fine-tuning overhead and achieves zero-shot generalization to unseen CSI sizes.

We study the achievable rate performance of dual-polarized passive reconfigurable intelligent surfaces (RISs) and dual-polarized network-controlled repeaters (NCRs) in a point-to-point wireless link. While RISs rely on passive beamforming with large array gains, NCRs exploit active amplification across polarization branches. We develop a unified analytical framework that captures the impact of amplification, noise propagation, and deployment geometry under different power-budget models. Closed-form rate expressions are derived, and the optimal one-stream and two-stream transmission strategies for NCRs are characterized. The results reveal that RISs are competitive in short-range scenarios and benefit from large array sizes, whereas NCRs become preferable in long-range deployments due to their ability to compensate for pathloss through amplification.

Cell-free massive MIMO with wireless fronthaul is a promising architecture for energy-efficient 6G networks, but the access and fronthaul links must then share the same scarce spectrum, and, under the fully centralized (option-8) functional split, the fronthaul rate is dictated by the finite quantization resolution used at the access points (APs). This paper develops a network energy-efficiency (EE) maximization framework for the uplink of such a system, jointly optimizing the integrated access and fronthaul (IAF) resource split, the adaptive per-AP quantization resolution, and the fronthaul powers, and treating the time-division (TD) and frequency-division (FD) operating modes in a unified manner. Each AP may be switched off (put to sleep) when it is not worth activating, so the resolution allocation is inherently coupled with AP selection. The resulting mixed-integer, nonconvex fractional program is solved by an alternating-optimization algorithm with per-block optimality guarantees—a closed-form optimal time split, bandwidth bisection, and optimal per-AP bit selection—that applies verbatim to both modes. While the design relies on the tractable additive quantization noise model, the reported performance is obtained end-to-end with the actual Lloyd–Max quantizers and a Bussgang decomposition-based achievable-rate bound.

In recent years, the term voiceprint has regained attention, particularly in technological applications and policy-making contexts, often carrying the assumption that a person's voice constitutes a stable and unique biometric trace analogous to a fingerprint. Yet this conception has been repeatedly criticized and rejected by forensic voice experts throughout the decades since its introduction. Although voices undoubtedly contain speaker-related information, this simplified conception obscures the highly dynamic and context-dependent nature of speech. This article revisits the voiceprint fallacy and reconsiders what can count as evidence of speaker identity by reviewing the historical development of voiceprint identification, evidence on human voice variability, developments in forensic voice comparison, research on human and automatic speaker recognition, and the recent challenge posed by deepfake speech to speaker identity. We point out that the voiceprint metaphor and its underlying implications are scientifically misleading because they transform a probabilistic source of speaker information into an imagined stable object of identity. We argue that speaker identity assessment does not require, and current evidence does not support, the existence of a stable and individually unique voiceprint. For speaker recognition and voice biometrics, this distinction motivates interpreting learned speaker representations with respect to the conditions under which they are trained and evaluated, and explicitly assessing their robustness to relevant sources of within-speaker variability, domain mismatch, and synthetic manipulation.

Neural synthesis for musical instruments has the potential to revolutionize current practices that use concatenative synthesis and a sample library. However, most research focused on piano synthesis and expressive performance generation; little work has been done on continuously articulated instruments like the violin, let alone rendering them with playing techniques and dynamics. We present VIOLET, a latent-diffusion framework for controllable violin synthesis, which uses a Diffusion Transformer (DiT) with rectified flow to synthesize high-fidelity audio from MIDI notes, playing techniques, and continuous dynamics. To train VIOLET, in addition to using a few existing datasets, we curate a new dataset named CSV-TD, which contains 39 h of 48 kHz synthetic audio and time-aligned annotations of MIDI notes, note-level techniques, and continuous dynamics curves. Objective and subjective evaluations show that VIOLET synthesizes violin performances with high technique adherence, accurate pitch and timing alignment, and good dynamics control. It outperforms the current state-of-the-art neural violin synthesis system and approaches a top commercial virtual instrument in terms of technique clarity, naturalness, and dynamics following.

In many electromagnetic systems, the set of radiated or scattered fields is controlled by only a few physical parameters and therefore forms a low-dimensional manifold embedded in a high-dimensional observation space. This paper extends Electromagnetic Information Theory (EIT) to such parametric field families by separating two descriptors that classical linear NDF analysis merges: the Electromagnetic Intrinsic Dimension (EID), $d$, which counts the locally independent directions of field variation, and the Metric Stretching Exponent, $\nu$, which governs the electrical-size scaling of the intrinsic metric volume. Using Kolmogorov $\varepsilon$-entropy and $\varepsilon$-capacity, we derive lower bounds showing that stable non-linear representations depend not only on dimension but also on metric-volume growth, which reappears as a decoder-sensitivity burden. Under additive Gaussian noise, the pullback metric is proportional to the Fisher Information Matrix, linking the same geometry to Cramér–Rao estimation bounds. Physics-constrained autoencoders provide an operational estimate of the latent dimension required to achieve a prescribed reconstruction accuracy and an empirical proxy for the associated normalized decoder sensitivity. Array and scattering benchmarks show that systems with the same intrinsic dimension can exhibit different metric-growth laws depending on the physical modulation or scattering regime, while a dedicated steering-arc experiment provides finite-sample evidence that the best observed decoder-sensitivity proxy scales nearly linearly with metric length across electrical apertures, consistently with the predicted lower-bound trend.

This article studies the computation of the maximal controlled invariant set (MCIS) for neural network control systems (NNCSs) with a nominal plant model and a neural network residual model. An interval-based method is developed to construct a control-affine inclusion function for the NNCS dynamics and to exploit the induced hyperplane arrangement in the input space, which reduces the verification of controlled invariance from an infinite control search to a finite set of representative evaluations. Based on this finite characterization, verification-guided algorithms iteratively construct inner and outer approximations of the the MCIS, which are naturally parallelizable. Numerical examples demonstrate the effectiveness of the proposed approach.

In LEO satellite networks utilizing beam hopping (BH), resource allocation plans must be committed well in advance. This inherent operational delay necessitates predicting future user demand during the planning phase. Such predictive agility is particularly crucial for military applications, where unpredictable tactical environments demand low-latency, resilient communication links. However, existing forecasting models are typically evaluated based on standalone accuracy, ignoring their cross-layer impact on overall network performance. To address this gap, we evaluate two distinct demand forecasting solutions within a comprehensive, full-stack LEO satellite simulation compliant with DVB-S2X standards. Beyond prediction accuracy, we examine how incorporating user demand forecasts into BH plan generation impacts key network metrics, particularly delay and jitter. We evaluate these forecasting solutions alongside a static allocation baseline. Our results demonstrate that forecast-based dynamic planning reduces delay by 10-40% across the beams under certain load conditions compared to static allocation methods. Crucially, marginal improvements in predictive accuracy do not translate into proportional network metric gains. While the evaluated forecasting solutions differ by 14-16% in Normalized Mean Square Error (NMSE), this discrepancy yields less than a 1% reduction in delay and produces nearly identical jitter characteristics. These findings suggest that when designing user demand forecasting solutions for practical LEO deployments, prioritizing system scalability may be more valuable than chasing minor accuracy enhancements.

Driven by advances in diffusion models and autoregressive models, the fidelity and resolution of AI-generated images now rival those of real images. However, existing AI-generated image detection methods often downsample the images,inevitably overlooking critical low-level texture details in high-resolution AI-generated images, thereby limiting their detection performance. In addition, the continuous emergence of unknown generative models makes large-scale pre-training datasets inaccessible. To address these challenges, we propose a novel AI-generated image detector, termed LHSDet (Low-, High-, and Semantic-level Detection). Specifically, we formulate the AI-generated image detection task as a Visual Question Answering problem, leveraging a fine-tuned vision-language framework to fully exploit the complementary information between visual and textual modalities. Recognizing that the default visual encoder of existing vision-language models is not tailored for AI-generated image detection, we redesign a visual encoder to better capture both the low-level and high-level artifacts inherent in AI-generated images. Furthermore, we incorporate a semantic-level textual branch to enable multi-modal feature fusion and detection. Consequently, LHSDet employs a triple-branch architecture to extract complementary multi-modal features: a low-level visual branch that aggregates non-overlapping patches for local texture cues, a high-level visual branch based on SigLIP2 for global perception feature extraction,and a semantic-level textual branch that generates captions using BLIP-2. Extensive experimental results demonstrate that LHSDet achieves strong detection accuracy and robust performance across diverse generative models.

In this paper, we highlight opportunities and challenges of deploying Rayleigh-based distributed optical fiber sensing, also known as distributed acoustic sensing (DAS), over existing optical telecommunication networks. We review some of the prominent recent advances in this area and we shed light on future directions. First, we briefly review classifications of DAS systems and we focus on the advantages of using a continuous-wave constant-power DAS probe signal for in-band coexistence between DAS and data transmission over the same fiber. Second, we emphasize on the important role of the mechanical coupling between the fiber, its hosting cable and their surrounding environment in the interpretation of the extracted strain information. Third, we show results on the estimation of distributed birefringence information using a polarization-diversity DAS interrogation. Finally, we study the impact of laser phase noise on continuous-wave constant-power DAS interrogation using phase-encoded probing sequences and attempt to mitigate it.

Large Audio Language Models (LALMs) perform well on complex question answering but often show a reasoning gap, where explicit Chain-of-Thought (CoT) reduces accuracy compared to direct answers. We hypothesize that long reasoning sequences shift attention away from the audio input. To address this, we propose SPARE (Semantic Prediction for Audio REasoning), which introduces a register token aligned with the final conclusion using a cosine similarity loss with a Sentence-BERT embedding. This conditions the model's latent space with the target semantic goal before reasoning begins. Experiments on MMAU and MMAR with SALMONN show improved zero-shot reasoning and stronger early attention to audio without additional inference cost.

Running vision-language navigation fully onboard an aerial robot is hard, since grounding, planning, and control must share limited compute and a single-stage error is difficult to isolate in flight. End-to-end aerial policies fuse these stages into one network, giving up the observability and safety checks a modular stack keeps available. We propose VLN on the Fly, an onboard stack that keeps grounding, planning, and control as separate, inspectable stages. A quantized VLM grounds an instruction to a coarse image cell, depth lifts it to a 3D goal, a fast B-spline planner returns a feasible trajectory, and a pretrained reinforcement learning policy tracks it to motor commands across quadrotors. Across 15 onboard flights over three everyday referents in a controlled indoor volume, the stack reaches the target in 13 of 15 trials with 5.72 cm mean goal error and 39.3% average GPU utilization. In 6 additional cluttered-environment trials, the stack tracks collision-free trajectories under onboard perception gating.