Audio and Speech Processing · Cross-listed from cs.CV
Benchmarking Single-Factor Physical Video-to-Audio Generation
Tingle Li, Siddharth Gururani, Kevin J. Shih et al. (+6 more)
Generative
video-to-audio (V2A) models produce highly plausible soundtracks, but it remains unclear whether they capture the underlying physical processes. Existing evaluations emphasize perceptual realism and overlook physical correctness under controlled interventions. In this paper, we introduce FlatSounds, a
benchmark that audits the physical
reasoning of V2A models through: 1) controlled counterfactual pairs in which a single physical factor is varied, and 2) single-
video pattern tests that probe internal consistency and directional trends. These settings test whether the generated audio correctly reflects specific physical properties and timings. Our evaluation of state-of-the-art models reveals a consistent trade-off: models rely more on text captions than the visual stream to infer physics and semantics. Captions generally improve physical and semantic accuracy, but paradoxically degrade temporal
alignment. Our results highlight the need to move beyond audio quality toward learning physical processes directly from pixels. Finally, we find that our physics-based metrics correlate strongly with human preference tests on our own data. Project webpage: https://research.nvidia.com/labs/cosmos-lab/flatsounds/
Read the full abstract ▾
All authors Tingle Li, Siddharth Gururani, Kevin J. Shih, Gantavya Bhatt, Sang-gil Lee, Zhifeng Kong, Arushi Goel, Gopala Anumanchipalli, Ming-Yu Liu
Image and Video Processing · Cross-listed from cs.CV
Boosting Image Quality Assessment Performance: Unsupervised Score Fusion by Deep Maximum a Posteriori Estimation
Zhongling Wang, Raymond Zhou, Shahrukh Athar et al. (+2 more)
Over the past decades, numerous Image Quality Assessment (IQA) models have emerged, aiming to predict the perceptual quality of images. However, individual models are often biased toward certain types of image content or distortions, depending on the design principle and process. An intuitive idea is to harness the strengths and mitigate the weaknesses of each IQA model, by fusing the scores of multiple models into a stronger one. Here we make one of the first attempts to seek an optimal solution for the idea and propose a general framework for unsupervised IQA score fusion using deep Maximum a Posteriori (MAP) estimation. The proposed model conducts fine-grained uncertainty estimation at the score level to increase the accuracy and reduce the uncertainty in fused predictions. Comprehensive experiments demonstrate the superiority of the proposed model over individual IQA models and other fusion methods. It also exhibits an interesting capability of rejecting “bad" models in the fusion process.
Read the full abstract ▾
All authors Zhongling Wang, Raymond Zhou, Shahrukh Athar, Wenbo Yang, Zhou Wang
Systems and Control
Optimization of Predictive Maintenance Schedules under Uncertainty: A Scenario-Based Theoretical Framework
Jerzy Baranowski and Waldemar Bauer
This paper proposes a scenario-based framework for predictive maintenance scheduling under uncertainty in a finite planning horizon. The considered setting involves multiple assets for which maintenance decisions are informed by three heterogeneous sources of information: calendar-based overhaul intervals, usage-based limits driven by uncertain future operating cycles, and condition-monitoring outputs represented through remaining useful life (RUL) estimates with uncertainty. While these elements have been studied extensively in the maintenance literature, they are often treated separately or only partially integrated. In contrast, the proposed formulation evaluates complete maintenance schedules under simulated future scenarios and compares them using expected-cost and tail-risk criteria. The contribution is primarily conceptual and methodological: we define a unified finite-horizon decision framework that combines calendar-, usage-, and prognostics-based information within a common scheduling problem. A small synthetic computational example is used as a proof of concept. The results show that integrated scenario-based policies can substantially outperform simpler single-trigger rules, while the difference between risk-neutral and risk-aware integrated policies remains modest under the present calibration.
Read the full abstract ▾
Systems and Control
Fault-Ride-Through Coordination Strategy for Offshore AC Islands with Multi-Infeed HVDC Interconnections
Eleni Tsotsopoulou, Vasileios Psaras, Dionysios Moutevelis et al. (+2 more)
Large-scale offshore Wind Farms (WFs) are considered key assets towards realizing a sustainable power system. These systems are often configured as offshore AC islands and their integration largely depends on the High-Voltage-Direct-Current (HVDC) technology. This
topology, while it enables cost-effective transmission over large offshore distances, may lead to operational challenges. Specifically, the operation of offshore AC islands during faults and the grid code requirement fulfillment are identified as a major challenges for their large-scale deployment. To address this pressing issue, a comprehensive coordination control strategy for the different participating converters in multi-infeed AC offshore islands during Fault Ride Through (FRT) operation is presented in this work. The proposed strategy introduces advanced control functions in the FRT schemes of both the HVDC and WF converters, such as zero active and reactive power injection during faults, as well as post-fault active power droop control coordination to tackle power imbalances. The proposed FRT coordination strategy is validated through both extensive simulations in PSCAD/EMTDC, as well as with Power Hardware-in-the-Loop (PHIL) experimental results, considering both AC and DC faults.
Read the full abstract ▾
All authors Eleni Tsotsopoulou, Vasileios Psaras, Dionysios Moutevelis, Oriol Gomis-Bellmunt, Alexandros Paspatis
Signal Processing · Cross-listed from cs.CE
A Lumped-Element Electrical Model of the Human Head for Brain-Oriented Applications
Angelo Faccia, Ermanno Citraro and Francesco P. Andriulli
In this work, we present a compact surrogate
circuit for electro-quasi-static (EQS) head modeling. A three-shell geometry (
brain, skull, scalp) is considered, and each layer is modeled through radial and tangential pathways, implemented as RC branches. Frequency-dependent tissue conductivity and permittivity are mapped into dispersive resistive and capacitive elements. The model is validated against a semi-analytical spherical-harmonics reference solution over multiple geometrical configurations and operating frequencies, demonstrating good agreement. Neglecting dispersion and capacitive pathways can lead to an overestimation of scalp potentials over the considered frequency range, highlighting the need for dispersive RC
circuit modeling.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.HC
REACT: A Conditioning Framework for User-Adaptive sEMG Hand Pose Estimation
Eric Xie and Hei Shing Cheung
Surface electromyography (sEMG) enables continuous hand pose estimation on wearable devices, but models trained on multi-user corpora degrade on unseen individuals due to inter-user variability in anatomy and electrode placement. We propose REACT, a lightweight conditioning framework that personalizes a frozen pretrained EMG-to-pose backbone at inference time using only a handful of calibration recordings. REACT learns a compact user
embedding from calibration data and applies Feature-wise Linear Modulation (FiLM) to adapt the shared encoder's feature space, requiring no gradient updates at deployment. On the large-scale EMG2POSE
benchmark, REACT improves over the state-of-the-art baseline across all three generalization splits in both regression and tracking modes, reducing angular error by up to 3.9% with minimal parameter overhead and under 45 seconds of per-user calibration.
Read the full abstract ▾
Signal Processing · Cross-listed from math.ST
The generalized method of moments is (almost) statistically efficient in low-SNR Gaussian latent-variable models
Amnon Balanov, Tamir Bendory and Dan Edidin
We study estimation in the low signal-to-noise ratio (SNR) regime for a broad class of Gaussian latent-variable models, including Gaussian mixtures and orbit recovery problems. We show that, in this regime, the generalized method-of-moments (GMoM) matches the first-order asymptotic efficiency of maximum likelihood. In particular, if the moment features are chosen up to the minimal local order required for identification and are weighted optimally, then the resulting GMoM
estimator has the same leading asymptotic covariance as the maximum-likelihood
estimator. Our analysis shows that, in low SNR, this equivalence is governed by a layered local geometry: different directions become informative at different moment orders, partitioning the space into layers with distinct SNR scalings. We prove that the observed Fisher information and the GMoM information operator admit matching layerwise expansions across these layers. As a consequence, in the low-SNR regime, GMoM provides a statistically efficient alternative to maximum likelihood, while preserving the computational advantages of moment-based estimation.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.CE
A Lumped RC Equivalent Circuit of Head Tissues for Dispersive Neuro-Electromagnetic Modeling
Angelo Faccia, Ermanno Citraro and Francesco P. Andriulli
Accurate modeling of electric potential and current distribution in head tissues is crucial for the design and evaluation of neuro-sensing and neuro-stimulation systems operating in the sub-megahertz frequency range. Numerical methods are widely employed in electromagnetic simulations, however their computational cost can limit their applicability to rapid prototyping, real-time simulations, and
circuit-level integration. In this work, we introduce a lumped RC equivalent
circuit model that reproduces the electrical behavior of a canonical three-layer spherical head geometry over a frequency range up to 50 kHz. The model accounts for frequency-dependent tissue conductivity and permittivity to capture dispersive effects, employing complex conductivity in the electro-quasi-static (EQS) regime. The
circuit topology uses a minimal set of impedance elements in order to represent the essential mechanisms of electric signal propagation. Validation was performed using a dipolar
brain source configuration for scalp voltage peak estimation, showing close agreement with semi-analytical solutions across different skull thicknesses and dipole eccentricities. In addition, the impact of tissue dispersion and capacitive branches on the model predictions was quantitatively assessed, showing their contribution to the overall fidelity of the proposed approach.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.IT
Low-Overhead Receiver Design for Data-Dependent Superimposed Training via Deep Learning
Xinjie Li, Xingyu Zhou, Jing Zhang et al. (+3 more)
Superimposed pilot (SIP) transmission improves spectral efficiency by eliminating the dedicated pilot overhead required in orthogonal pilot (OP)-based schemes. However, SIP suffers from severe pilot-data coupling, which leads to a critical performance-complexity bottleneck at the receiver. To address this issue, this paper proposes a low-overhead transmission framework that revitalizes data-dependent superimposed training (DDST) with enhanced interference mitigation strategies. First, for quasi-static block-fading channels, an enhanced DDST receiver is developed to achieve non-iterative pilot-data decoupling by exploiting data-dependent algebraic structures. Second, to overcome the sensitivity of conventional DDST to channel variations and symbol misidentification in fast time-varying environments, a mix transmission scheme is developed. By strategically applying DDST to a subset of resource elements, the proposed scheme combines the interference-free transmission property of OP with the zero-pilot-overhead advantage of SIP, thereby improving demapping reliability and interference suppression. Furthermore, under the proposed mix scheme, a
Vision Transformer-based neural receiver is designed to capture the orthogonal structure between pilots and perturbation-bearing data, as well as the underlying channel correlations, thereby relaxing the stringent quasi-static assumption required for interference disentanglement. Simulation results demonstrate that the proposed framework achieves significant performance gains in the low-to-medium SNR regime under time-varying channels while providing superior computational efficiency compared with state-of-the-art SIP receivers.
Read the full abstract ▾
All authors Xinjie Li, Xingyu Zhou, Jing Zhang, Chao-Kai Wen, Xiao Li, Shi Jin
Signal Processing · Cross-listed from cs.LG
A Fully Convolutional Approach to Denoising 2D Correlation Spectra
Nisar Nellikunnummel, Andi M Barbour, Lutz Wiegart et al. (+2 more)
We present a fully convolutional denoising autoencoder (FC-DAE) tailored for two-dimensional representations of dynamic correlations that is applicable to many experimental techniques. Here, we demonstrate its performance on two-time intensity correlation functions ($C_2$) from X-ray
photon correlation spectroscopy (XPCS). Unlike conventional denoising autoencoders that are typically restricted to fixed input sizes, the FC-DAE accepts inputs of arbitrary dimensions while preserving correlation structures across diverse dynamical regimes. The model is trained using experimentally derived $C_2$ data collected at NSLS-II beamlines, with data augmentation applied to expand the diversity of the
dataset and reduce overfitting. The FC-DAE successfully recovers intricate dynamical features in low signal-to-noise conditions while maintaining structural fidelity. To assess reconstruction reliability, we employ quantitative metrics to evaluate structural fidelity and identify potential model-induced bias. Our results demonstrate that the FC-DAE provides robust denoising performance with high computational efficiency, enabling recovery of XPCS dynamics under
photon-limited and low-dose measurement conditions.
Read the full abstract ▾
All authors Nisar Nellikunnummel, Andi M Barbour, Lutz Wiegart, Tatiana Konstantinova, Anthony M DeGennaro
Audio and Speech Processing
Frequency-Modulated and Single-Tone Excitation to Reveal Vibro-Acoustic Nonlinearities in Loosened Bolted Joints
Berkay Kullukcu, Robin Pianowski and Dina Hannebauer
Preload loss in bolted joints results in alterations of the stiffness, damping, and nonlinearity of the structure, but existing monitoring techniques for rail-vehicle systems are often not capable of combining controlled shaker tests and sensing of nonlinear features. This paper proposes a method for detecting bolt loosening using a vibro-acoustic technique, where the structure is subjected to controlled shaker tests to sense the nonlinear features. A triaxial accelerometer was attached to the demonstrator, a microphone was placed in close proximity, and one of the bolts was tested under 0%, 20%, 40%, and 80% preload conditions. Single-tone and frequency-modulated (FM) signals close to the main natural frequency of 130 Hz, which was identified using sine sweep and narrow-band excitation, were applied to the demonstrator. When the structure was subjected to 130 Hz single-tone excitation, the loose state of the bolt exhibited several additional high-frequency spectral peaks. FM excitation between 125 and 135 Hz further distinguished between the states. Harmonic band power ratios, normalized to the carrier, distinguished between the loose state and the 80% preload state, where the difference between the loose and 80% preload states was 17.5 dB for l = 2 and 36.5 dB for l = 6.
Read the full abstract ▾
Audio and Speech Processing · Cross-listed from cs.SD
HoliTok: A Continuous Holistic Tokenization with Robust Dual Capabilities of Speech Generation and Understanding
Bohan Li, Shi Lian, Hankun Wang et al. (+6 more)
Unified
speech foundation models require a holistic tokenization space that is both learnable by
language models and decodable into high-quality waveforms. Existing
speech tokenizers, however, often fail to satisfy these requirements simultaneously, leading to increased architectural complexity and more involved training designs. We propose HoliTok, a continuous Holistic
speech Tokenization model designed for unified generation-understanding modeling. HoliTok encodes 48~kHz
speech into a compact 25~Hz sequence of 128-dimensional latents. It is trained with a progressive strategy that jointly preserves signal-level fidelity, incorporates semantic information, and maintains strong latent learnability. Based on this tokenization, we build a unified AR+DiT model for
speech synthesis and recognition, where the same latent sequence supports both generation-specific and unified generation-understanding tasks. Experiments show that HoliTok achieves competitive reconstruction fidelity, improves generative learnability for high-quality and controllable synthesis, and, among the evaluated representations, is the only one that operates robustly in our unified generation-understanding architecture without additional optimization tricks. These results suggest that HoliTok serves as an effective
speech tokenizer and a foundational representation interface for unified spoken
language modeling. The code is available at: https://github.com/bovod-sjtu/HoliTok.
Read the full abstract ▾
All authors Bohan Li, Shi Lian, Hankun Wang, Yiwei Guo, Yu Xi, Zhihan Li, Da Zheng, Colin Zhang, Kai Yu
Image and Video Processing · Cross-listed from physics.app-ph
Reconfigurable Multistate MRAM Synapses with Vortex STNO based Neurons for Scalable In-Memory Convolutional Neural Networks
Ravish Kumar Raj, Simon N. Richter, Saeed Baghaee Ivriq et al. (+9 more)
Magnetic tunnel junction (MTJ)-based
magnetic random-access
memory (MRAM) is a promising platform for neuromorphic and in-
memory computing owing to its non-
volatility, high endurance, fast switching dynamics and CMOS compatibility. However, conventional
spin-transfer torque and
spin-orbit torque MRAM implementations for neural networks often suffer from high critical switching currents, large latency, thermal instability and significant read-write overheads. Here, we demonstrate a unified multistate MRAM-
spin-torque nano-oscillator (STNO) architecture that integrates synapses and
neurons on a single chip for convolutional neural network (CNN) applications. The system employs 1x8 multistate MRAM arrays as programmable synapses coupled with a
vortex-based STNO
neuron, enabling both individual and collective programming through fieldline-driven write channels. Multiple configurable resistance states are achieved by tuning internal and external
magnetic fields together with bias currents, allowing quantized positive and negative synaptic weights for configurable kernel and pooling operations. The proposed architecture is evaluated through simulation on MNIST, SVHN, CIFAR-10, Google
Speech Commands (GSC) and RadioML
datasets, achieving accuracy of 99.76%, 87.93%, 78.14%, 87.96% and 56.46% respectively. Based on fabricated device dimensions, the complete architecture occupies ~6171.2 {\mu}m2 with an average energy consumption of 200.08 pJ per training and inference cycle for MNIST, highlighting its potential for scalable low-power neuromorphic computing
Read the full abstract ▾
All authors Ravish Kumar Raj, Simon N. Richter, Saeed Baghaee Ivriq, Oliver Fridorf, Darío Fernández-Khatiboun, Yasser Rezaeiyan, Luana Benetti, Tim Boehnert, Ricardo Ferreira, Hooman Farkhani, Sonal Shreya, Farshad Moradi
Audio and Speech Processing · Cross-listed from cs.AI
It`s All About Speed: AI`s Impact on Workflow in Music Production
Finn McClellan and Fabio Morreale
In this paper, we present the results of an ethnographic study into the impact of AI and automated tools on music production workflow. Focusing specifically on professional participants who identified as recording engineers, mixers, and producers, we discuss their usage of common AI and automated software, as well as their sentiments on the proliferation of these tools. We discuss tensions that may be created between users and automated tools in key areas such as the need for speed and efficiency, controllability, and maintaining creative agency, and how these tensions may be alleviated through tool design.
Read the full abstract ▾
Audio and Speech Processing
Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions
Heejoon Koo, Yoon Tae Kim, Miika Toikkanen et al. (+1 more)
AI-driven respiratory sound classification (RSC) is promising for automated pulmonary disease detection, yet multi-site deployment is hindered by inter-stethoscope variability. We introduce a
federated domain generalization (FedDG) formulation for RSC in which clients hold recordings from different stethoscopes and the model is evaluated on an unseen device. Our empirical analysis shows that stethoscope-induced style and disease-relevant content are partially entangled, making deterministic style removal unreliable. In response, we propose BTS-CAFE, a framework combining (i) causality-inspired device-style interventions with constraints designed to limit content distortion, (ii) counterfactual metadata augmentation to relieve device and demographic shortcuts, and (iii) gradient
alignment to promote device-invariant decision boundaries across clients. Built on BTS with CLAP, a
multimodal language-audio pretraining model, BTS-CAFE improves the out-of-distribution ICBHI Score by 3.69 points on average over five held-out devices relative to its backbone, and outperforms conventional data augmentation and
federated learning baselines in simulated device leave-out evaluations on the ICBHI and SPRSound
datasets.
Read the full abstract ▾
All authors Heejoon Koo, Yoon Tae Kim, Miika Toikkanen, June-Woo Kim
Audio and Speech Processing
MELD: Mel-Spectrogram-Based Speech Language Modeling with Discrete Latent Variables
Sung-Lin Yeh, Wei Zhou, Gil Keren et al. (+6 more)
Recent
speech language models rely on encoders that are optimized separately from autoregressive models. Since these encoders are unaware of the downstream objectives, the extracted representations may not be optimal for downstream tasks. To address this limitation, we introduce a discrete latent variable model on mel spectrograms that jointly optimizes the encoder and the
speech language model. Joint optimization not only brings improvements over codec-based and other mel-spectrogram-based baselines on zero-shot Text-to-
Speech (TTS) and
Speech-to-Text (STT) tasks, but also effectively alleviates common issues in autoregressive mel spectrogram modeling, such as prolonged silence generation and word omissions.
Read the full abstract ▾
All authors Sung-Lin Yeh, Wei Zhou, Gil Keren, Duc Le, Zhong Meng, Hao Tang, Jay Mahadeokar, Ozlem Kalinli, Alexandre Mourachko
Systems and Control
BuilDyn: Excitation-Driven Data Generation for Building Thermal Dynamics Modeling and Control
Felix Koch, Thomas Krug, Fabian Raisch et al. (+2 more)
Machine learning (ML) is increasingly used for data-driven modeling of buildings to enable downstream tasks such as fault detection and diagnosis, and energy-efficient control. While recent work improves generalization across building characteristics, weather, and occupancy, generalization also depends on sufficient exploration of the control-driven system state space. Existing real-world
datasets and simulation environments predominantly reflect stationary operation under fixed control policies, resulting in limited excitation and reduced robustness to unseen operating conditions. This paper introduces BuilDyn, a package based on BuilDa that enables customizable excitation strategies for control-oriented data generation. BuilDyn further supports sampling from representative building distributions and provides a Python interface for easy integration into machine learning pipelines. We demonstrate the benefits of BuilDyn by comparing the performance of data-driven ML models trained on non-excited and excited data for one building. With BuilDyn, we hope to advance scalable control-oriented modeling and support future directions such as transfer learning and building-specific foundation models.
Read the full abstract ▾
All authors Felix Koch, Thomas Krug, Fabian Raisch, Benjamin Schäfer, Benjamin Tischler
Systems and Control
Distributed Nonlinear Model Predictive Control for District Heating Networks
Alessandro Bettoni, Giacomo Mastroddi and Marco Muttoni
This paper presents a distributed nonlinear model predictive control that uses alternating direction method of mul tipliers for district heating networks. Exploiting a graph-based modeling of the thermal dynamics, our controller optimizes the mass flow absorption of buildings in a distributed cooperative scheme that mediates between the superior performance of the centralized control and the
privacy preservation of the decentralized schemes. A
benchmark three-building network simulation is used to compare the performance of the proposed solution with a decentralized model predictive control scheme.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.IT
On the Effect of Pulse Shaping Filters in Zak-OTFS Waveform for Radar Sensing
Abhishek Bairwa and Ananthanarayanan Chockalingam
In radar sensing, the self-ambiguity function of the probing waveform plays a crucial role in the resolvability and detection of multiple targets. In the recent Zak-OTFS based radar literature, Gaussian pulse shaping filter has been considered, and it has been shown to offer better range/velocity estimation performance compared to the traditional chirp waveform in scenes with multiple targets. While the self-ambiguity function with Gaussian filter has very low side lobes, its main lobe is wide which compromises resolvability and performance. Motivated by this, we seek filters with better ambiguity characteristics. Specifically, we explore two other known filters, namely, sinc and Gaussian-sinc (GS) filters, and demonstrate that these filters offer better performance compared to Gaussian filter under different scenarios and receiver processing. Towards demonstrating this, we derive closed-form expressions for the self-ambiguity functions of Zak-OTFS waveform with sinc and GS filters. The ambiguity functions of sinc and GS filtered waveforms have narrow main lobes, resulting in better resolvability in scenes with densely populated targets for the basic peak-detection based receiver. The ambiguity function of Gaussian filtered waveform has very low sidelobes, resulting in better performance in sparsely populated scenes. When a receiver with inter-target interference mitigation is used, the sinc and GS filters perform better in both dense and sparsely populated scenes compared to Gaussian filter.
Read the full abstract ▾
Systems and Control
Teleoperation Operational Design Domain based on Minimal Risk Maneuver Capability
Leon Johann Brettin, Nayel Fabian Salem, Ole Hans et al. (+1 more)
This article discusses the concept of an Operational Design Domain (ODD) designed specifically for teleoperated road vehicles. For this purpose, the ODD concept designed for automated driving is adapted for teleoperation. As teleoperation becomes more common in regular traffic, the question arises under which operating conditions such vehicles are able and allowed to drive. Currently, these conditions are selected primarily based on network performance. From a safety perspective, it is difficult to base such a selection on a reliable connection because it is almost impossible to guarantee sufficient reliability. With this in mind, the ODD concept designed for automated driving is adapted for teleoperation: A concept is proposed for basing the ODD for a teleoperation system on the capability of the teleoperated vehicle to perform a minimal risk maneuver using a dedicated system designed solely for this purpose. This concept is then demonstrated using a use case example.
Read the full abstract ▾
All authors Leon Johann Brettin, Nayel Fabian Salem, Ole Hans, Markus Maurer
Systems and Control · Cross-listed from cs.IT
Tackling Interference in HAPS Networks via Angular-Aware Clustering and RSMA
Afsoon Alidadi Shamsabadi, Animesh Yadav and Halim Yanikomeroglu
High Altitude Platform Stations (HAPS) have emerged as a promising enabler for next-generation wireless networks, offering ubiquitous connectivity to ground users. Operating either in standalone mode or in integration with terrestrial networks, HAPS can significantly enhance both coverage and capacity due to their strategic placement in the stratosphere. However, interference management in HAPS-empowered networks requires special
attention due to the unique propagation characteristics of HAPS links. In particular, the strong line-of-sight (LoS) conditions between HAPS and ground users result in limited channel variability, thereby intensifying inter-user interference. In this work, we consider a single HAPS serving multiple ground users through multiple beams over a limited number of orthogonal resource blocks (RBs). To address the resulting interference, we propose a novel angular-aware user clustering and interference-aware RB allocation framework that strategically clusters users, designs beams to serve each cluster, and allocates RBs to users across clusters. To further mitigate intra-RB interference, a rate-splitting multiple access (RSMA) scheme is incorporated. Simulation results demonstrate that the proposed clustering and RSMA-based approach significantly outperforms baseline schemes in terms of achievable per-user spectral efficiency.
Read the full abstract ▾
Image and Video Processing
Absorption and Phase-Contrast Microtomography Using Direct X-Ray Detection With Commercial Off-The-Shelf CMOS Sensors
Damian L. Corzi, Jose Lipovetzky, Fabricio Alcalde Bessia et al. (+5 more)
This work presents high-resolution X-ray microtomography based on commercial off-the-shelf (COTS) CMOS image sensors as direct detectors. The system uses a microfocus X-ray source in cone-beam geometry, enabling both absorption-contrast imaging, thanks to the sensor s intrinsic resolution, and phase-contrast imaging, by adjusting the object-to-sensor distance. A dynamic flat-field correction
algorithm mitigates radiation-induced sensor degradation during long acquisitions. The setup provides voxel sizes from 3.9 microns to 5.2 microns, with an effective spatial resolution (FWHM) as fine as 9.6 microns in magnified phase-contrast mode (1.33x) and 21 microns in non-magnified absorption-contrast mode. Phase contrast enhances visualization of soft tissue boundaries that are difficult to visualize using absorption-contrast imaging in the same experimental configuration.
Read the full abstract ▾
All authors Damian L. Corzi, Jose Lipovetzky, Fabricio Alcalde Bessia, German Mato, Andres Cicuttin, Maria L. Crespo, Martin Perez, Mariano Gomez Berisso
Image and Video Processing · Cross-listed from cs.CV
Low-Magnification SEM May Suffice: Interpretable Deep Learning for Multi-Scale Fracture-Cause Classification in Zirconia-Toughened Alumina
Julian Schmid, Pawel Astankow, Tom Vater et al. (+3 more)
Reliable identification of fracture origins in alumina
matrix composite hip and knee implants is critical for quality assurance and patient safety, yet current fractographic workflows are time-consuming, partly subjective, and reliant on high-magnification scanning electron microscopy (SEM). We present an interpretable
vision-
transformer (ViT) workflow for automated classification of fracture causes in an alumina
matrix composite (BIOLOX delta, CeramTec GmbH) widely used in total joint replacements. A
dataset of 8,493 SEM images (50x-10,000x) was curated from five years of in-production burst and proof tests and annotated into three defect categories defined along the manufacturing chain: green body, hard machining, and material defects. Under severe class imbalance, the fine-tuned ViT reached an accuracy of 0.907 and a macro-F1 of 0.888 in stratified five-fold cross-validation, with a two-stage perceptual-hash/SSIM leakage audit confirming negligible specimen overlap. Notably, performance at low magnification (50x) was comparable to that at high magnification (1k-10kx), indicating that macro-scale features - mirror geometry and hackle line fields - already encode sufficient diagnostic signal. Grad-CAM attributions consistently localised on canonical fractographic cues (mirrors, hackles, pores, machining marks), aligning with established fractographic criteria. Together, these results position interpretable ViTs as a complementary tool for ceramic-implant quality assurance, enabling low-magnification pre-screening and reducing reliance on time-intensive high-magnification inspection.
Read the full abstract ▾
All authors Julian Schmid, Pawel Astankow, Tom Vater, Julius Beck, Robert Cichon, Danny Krautz
Signal Processing
Multi-Snapshot Deep Denoising for Channel Estimation in OTFS Modulated Systems
Surbhi Gehlot, Siddhi Shinde, Suraj Srivastava et al. (+1 more)
A deep denoising based channel estimation framework is proposed for orthogonal time frequency space (OTFS) modulated systems, wherein channel state information (CSI) recovery is formulated as an image restoration problem. A salient attribute of the approach is the exploitation of structural invariance in the delay Doppler (DD) domain channel over a
geometric coherence time, allowing multiple OTFS frames captured during this period to serve as noisy snapshots of the approximately identical channel. These snapshots jointly enhance the effectiveness of the proposed lightweight denoiser based on nonlinear activation free network (NAFNet). The method exhibits low computational complexity, operates reliably even at low pilot signal-to-noise ratio (PSNR), and can accommodate both fractional delay and fractional Doppler effects. Simulation results demonstrate significant performance gains over the existing methods.
Read the full abstract ▾
All authors Surbhi Gehlot, Siddhi Shinde, Suraj Srivastava, Sandeep Kumar Yadav
Image and Video Processing
A unified deeplearning framework for contrast-phase-specific virtual monochromatic imaging
Antony Jerald, Hemant K Aggarwal, Brian Nett et al. (+4 more)
Dual-energy CT (DECT) enables virtual monochromatic imaging (VMI) and improved contrast resolution, but its clinical adoption is limited by hardware complexity and cost. In this work, we propose a unified deep learning framework that synthesizes contrast-phase-specific virtual monochromatic 50 keV images from single-energy CT (SECT) data by leveraging contrast phase information as a prior. The model is trained using DECT-derived 70 keV and 50 keV image pairs across four contrast phases – Angio, Arterial, Portal, and Delayed – using a novel prior conditioning architecture that integrates contrast phase priors into the energy transformation process. We demonstrate that the proposed unified model achieves contrast enhancement and generalizes well across contrast phases. Additionally, we show that the model can generate 50 keV-like images from SECT inputs, preserving contrast phase-specific dynamics.
Read the full abstract ▾
All authors Antony Jerald, Hemant K Aggarwal, Brian Nett, Avinash Gopal, Phaneendra K Yalavarthy, Bipul Das, Rajesh Langoju
Signal Processing · Cross-listed from cs.IT
A Unified Two-Stage Generative Diffusion Framework for Channel Estimation and Port Selection in Multiuser MIMO-FAS
Erqiang Tang, Wei Guo, Hengtao He et al. (+3 more)
Fluid
antenna systems (FAS) have emerged as a promising technology for next-generation wireless systems. However, practical multiuser multiple-input multiple-output FAS (
MIMO-FAS) faces two inherently coupled challenges: acquiring accurate high-dimensional channel state information (CSI) from limited RF chains and solving the combinatorial port selection problem, where the effectiveness of the latter highly depends on the result of the former. In this paper, we propose a unified two-stage
diffusion framework that formulates the joint task as a maximum-a-posteriori (MAP) inference problem equivalent to maximizing the posterior expected utility of the port selection, and approximates it through two sequential stages. For Stage I, we employ a continuous flow-based
diffusion model as a powerful implicit prior for 2D FAS channels and perform approximate posterior sampling via a parallel guided generation scheme, enabling accurate multiuser channel recovery even under severely low sub-sampling ratios. For Stage II, we train a discrete
diffusion model on optimal labels computed by exhaustive search to propose a pool of promising port selection candidates. The final selection is then determined by maximizing the empirical posterior expected utility over this pool, followed by a lightweight local refinement. Extensive simulations demonstrate that the proposed framework simultaneously achieves accurate channel estimation and delivers over 90% of the deployed rate of the computationally expensive exhaustive search over the estimated channel, while substantially improving the minimum achievable rate relative to practical cascaded schemes.
Read the full abstract ▾
All authors Erqiang Tang, Wei Guo, Hengtao He, Shenghui Song, Jun Zhang, Khaled B. Letaief
Signal Processing · Cross-listed from cs.HC
Embodied Virtual Reality Feedback Reshapes Neural Representations to Support Continuous Three-Dimensional Motor Imagery Decoding
Niall McShane, Attila Korik, Karl McCreadie et al. (+3 more)
Continuous
brain-computer interfaces (BCIs) that decode motion trajectories from imagined movement offer intuitive motor control, yet how feedback modality and longitudinal training shape neural representations and decoding performance remains poorly understood. We present the first systematic investigation of
embodied virtual reality (VR) feedback during real-time 3D virtual limb control driven by motor imagery, across ten longitudinal sessions in ten participants. Performance was evaluated using three strategies: actual online performance (Fixed Decoder Generalisation, FDG), periodic retraining (Sequential Adaptive Training, SAT), and within-session upper-bound estimation (Within-Session Reconstruction, WSR). A CNN-LSTM decoder achieved within-session imagined movement correlations of r = 0.762 under VR and r = 0.672 under screen feedback. VR significantly outperformed screen feedback across all strategies and movement dimensions (improvements of 8.9-13.0%, all p <= 0.002, d = 1.42-2.05). This advantage persisted under fixed decoders without retraining, demonstrating that
embodied VR feedback elicits inherently more decodable and generalisable neural representations. Linear mixed-effects modelling confirmed robust main effects of feedback modality and movement axis with no interaction. Neurophysiologically, VR produced stronger sensorimotor-parietal desynchronisation and enhanced motor-frontal functional connectivity, with pervasive anterior insula engagement across all frequency bands and increased superior parietal lobule coupling, paralleling patterns associated with real movement execution. These findings establish
embodied spatial feedback as a key design principle for next-generation continuous BCIs targeting intuitive motor control and neurorehabilitation.
Read the full abstract ▾
All authors Niall McShane, Attila Korik, Karl McCreadie, Naomi Du Bois, Darryl Charles, Damien Coyle
Audio and Speech Processing · Cross-listed from cs.SD
COMET: Concept Space Dissection of the Modality Gap in Audio-Text Multimodal Contrastive Embeddings
Yonggang Zhu, Liting Gao, Aidong Men et al. (+1 more)
Contrastive
Language-Audio Pretraining (CLAP) models are widely used for audio understanding and support modality-agnostic condition swapping in many zero-shot applications. However, their performance is heavily affected by the modality gap between audio and text
embeddings. Existing explanations mainly attribute this gap to the cone effect, treating it as a shift between mean
embeddings, yet correcting the mean alone yields only limited improvements. Alternative hypotheses, such as information imbalance and dimensionality collapse, have also been proposed, but they remain insufficiently verified and have not been thoroughly studied in the audio domain. Meanwhile, several works attempt to decompose
multimodal contrastive
embeddings into interpretable concepts, but none explicitly analyze the modality gap from the perspective of concept decomposition. In this work, we introduce COMET (Concept space Organization and Modality gap Explanation with PLS-SVD Transformation), a novel partial least squares singular value decomposition (PLS-SVD) framework for CLAP that unveils a broader perspective of the modality gap. Our framework reveals that only a small, interpretable subset of axes, which captures shared concepts, contributes substantially to similarity computation, and that the mean component represents only partially the modality gap. Building on this insight, we propose a simple spectral truncation method that mitigates the modality gap in a training-free manner. The method enables zero-shot audio captioning with condition swapping to approach fully supervised performance, without requiring large auxiliary
memory banks or expensive computation. At the same time, it achieves substantial
embedding dimensionality reduction while preserving strong performance on
retrieval and audio captioning tasks.
Read the full abstract ▾
All authors Yonggang Zhu, Liting Gao, Aidong Men, Wenwu Wang
Audio and Speech Processing
Decoding Strategies for Diffusion-Based ASR: A Systematic Evaluation of Confidence-Based Thresholding
Jeong Hun Yeo, Minsu Kim, Hyeongseop Rha et al. (+1 more)
While LLM-based Automatic
Speech Recognition (ASR) achieves high accuracy, its speed is limited by sequential autoregressive decoding.
Diffusion Language Models (DLMs) offer a parallel alternative, yet their decoding strategies remain under-explored in ASR contexts. This paper analyzes three decoding schemes for DLM-based ASR: fixed-number, static confidence threshold, and dynamic confidence threshold. We introduce a round-wise analysis of decoding progress using Negative Log-Likelihood-based uncertainty as a proxy for prediction reliability. Our results show that both threshold-based strategies provide a better accuracy-speed trade-off than fixed-number schemes. This behavior is associated with the more concentrated confidence distribution observed in the evaluated ASR settings: many
tokens reach high confidence early, enabling multiple
tokens to be committed in early decoding rounds while lower-confidence
tokens are deferred to later rounds. The static-threshold strategy achieves accuracy close to autoregressive decoding at lower decoding cost.
Read the full abstract ▾
All authors Jeong Hun Yeo, Minsu Kim, Hyeongseop Rha, Yong Man Ro
Systems and Control
Robustness Enhancement of Consensus Networks: the Optimal Memory Depth
Jiamin Wang, Jian Liu, Feng Xiao et al. (+2 more)
Understanding what governs collective robustness and how it can be enhanced remains a central pursuit in network science. This paper investigates the robustness of multi-
agent consensus networks, quantified by the $H_2$ performance metric, and delves into the enhancing effect of
agents' local
memory on it. Inspired by the hierarchical temporal structure of
memory observed in neuroscience, we focus on the role of
memory depth, which reflects the temporal features of
memory from recent to remote. Building on linear extrapolation, we propose a consensus protocol with single-step
memory and tunable
memory depth, derive the necessary and sufficient condition for achieving consensus, and show that the protocol exhibits an inheritable consensus property across
memory depths. Furthermore, analytical expressions for the $H_2$ performance metric, which depend on the
memory factor,
memory depth, coupling gain, and
Laplacian spectrum, are established. Under balanced usage of real-time and
memory information, we demonstrate that
memory at any accessible depth enhances $H_2$ performance, and the optimal
memory depth occurs at either the most recent or the most remote
memory, contingent upon certain parameter regions. Further detailed discussions are provided to clarify the broader implications of our findings.
Read the full abstract ▾
All authors Jiamin Wang, Jian Liu, Feng Xiao, Haibin Duan, Yuanshi Zheng