Systems and Control
String-Level Ground Fault Localization for TN-Earthed Three-Phase Photovoltaic Systems
Yuanliang Li, Xun Gong, Reza Iravani et al. (+3 more)
The DC-side ground fault (GF) poses significant risks to three-phase TN-earthed photovoltaic (PV) systems, as the resulting high fault current can directly damage both PV inverters and PV modules. Once a fault occurs, locating the faulty string through manual string-by-string inspection is highly time-consuming and inefficient. This work presents a comprehensive analysis of GF characteristics through fault-current analysis and a simulation-based case study covering multiple fault locations. Building on these insights, we propose an edge-AI-based GF localization approach tailored for three-phase TN-earthed PV systems. A PLECS-based simulation model that incorporates PV hysteresis effects is developed to generate diverse GF scenarios, from which correlation-based features are extracted throughout the inverter's four-stage shutdown sequence. Using the simulated
dataset, a lightweight Variational Information Bottleneck (VIB)-based localization model is designed and trained, achieving over 93% localization accuracy at typical sampling rates with low computational cost, demonstrating strong potential for deployment on resource-constrained PV inverters.
Read the full abstract ▾
All authors Yuanliang Li, Xun Gong, Reza Iravani, Bo Cao, Heng Liu, Ziming Chen
Signal Processing
OTFS-based Integrated Positioning and Communication Systems with Low-Resolution ADCs
Yueyi Yang, Zeping Sui, Zilong Liu et al. (+1 more)
This paper proposes a two-phase orthogonal time frequency space (OTFS)-based integrated positioning and communication (IPAC) framework under realistic low-resolution analog-to-digital converters (ADCs). In the uplink phase, the positioning signal is used to estimate channel parameters, which are subsequently used to determine the user's position. The spatial smoothing-multiple signal classification
algorithm is introduced to estimate the angle-of-arrival, whereas an iterative interference cancellation scheme is conceived for the remaining parameters' estimation. The corresponding Cramer-Rao lower bounds of channel parameters and user position are also derived. During the downlink communication phase, the estimated parameters are exploited to improve beamforming at the base station. Simulation results evaluate the impact of ADC quantizer resolutions. Specifically, it is shown that enhanced downlink bit error rate performance can be achieved with improved uplink positioning, while the use of low-resolution ADCs induces noticeable performance degradation in the OTFS-IPAC system.
Read the full abstract ▾
All authors Yueyi Yang, Zeping Sui, Zilong Liu, Leila Musavian
Systems and Control
Hierarchical JEPA Meets Predictive Remote Control in Beyond 5G Networks
Abanoub M. Girgis, Ibtissam Labriji and Mehdi Bennis
In wireless networked control systems, ensuring timely and reliable state updates from distributed devices to remote controllers is essential for robust control performance. However, when multiple devices transmit high-dimensional states (e.g., images or
video frames) over bandwidth-limited wireless networks, a critical trade-off emerges between communication efficiency and control performance. To address this challenge, we propose a Hierarchical Joint-
Embedding Predictive Architecture (H-JEPA) for scalable predictive control. Instead of transmitting states, device observations are encoded into low-dimensional
embeddings that preserve essential dynamics. The proposed architecture employs a three-level hierarchical prediction, with high-level, medium-level, and low-level predictors operating across different temporal resolutions, to achieve long-term prediction stability, intermediate interpolation, and fine-grained refinement, respectively. Control actions are derived within the
embedding space, removing the need for state reconstruction. Simulation results on inverted cart-pole systems demonstrate that H-JEPA enables up to 42.83 % more devices to be supported under limited wireless capacity without compromising control performance.
Read the full abstract ▾
Signal Processing
Adaptive Temporal Dynamics for Personalized Emotion Recognition: A Liquid Neural Network Approach
Anindya Bhattacharjee, Nittya Ananda Biswas, K. A. Shahriar et al. (+1 more)
Emotion recognition from physiological signals remains challenging due to their non-stationary, noisy, and subject-dependent characteristics. This work presents, to the best of our knowledge, the first comprehensive application of
liquid neural networks for
EEG-based
emotion recognition. The proposed
multimodal framework combines convolutional feature extraction,
liquid neural networks with learnable time constants, and
attention-guided fusion to model temporal
EEG dynamics with complementary peripheral physiological and personality features. Dedicated subnetworks are used to process
EEG features and auxiliary modalities, and a shared autoencoder-based fusion module is used to learn discriminative latent representations before classification. Subject-dependent experiments conducted on the PhyMER
dataset across seven emotional classes achieve an accuracy of 95.45%, surpassing previously reported results. Furthermore, temporal
attention analysis provides interpretable insights into
emotion-specific temporal relevance, and t-SNE visualizations demonstrate enhanced class separability, highlighting the effectiveness of the proposed approach. Finally, statistical analysis of temporal dynamics confirms that the network self-organizes into distinct functional groups with specialized fast and slow
neurons, proving it independently tunes learnable time constants and
memory dominance to effectively capture complex
emotion artifacts.
Read the full abstract ▾
All authors Anindya Bhattacharjee, Nittya Ananda Biswas, K. A. Shahriar, Adib Rahman
Systems and Control · Cross-listed from cs.NE
Curriculum-Learned Vanishing Stacked Residual PINNs for Hyperbolic PDE State Reconstruction
Katayoun Eshkofti and Matthieu Barreau
Modeling distributed dynamical systems governed by
hyperbolic partial differential equations (PDEs) remains challenging due to discontinuities and shocks that hinder the convergence of traditional physics-informed neural networks (PINNs). The recently proposed vanishing stacked residual PINN (VSR-PINN) embeds a vanishing-viscosity mechanism within stacked residual refinements to enable a smooth transition from the
parabolic to
hyperbolic regime. This paper integrates three curriculum-learning methods as primal-dual (PD) optimization, causality progression, and adaptive sampling into the VSR-PINN. The PD strategy balances physics and data losses, the causality scheme unlocks deeper stacks by respecting temporal and gradient
evolution, and adaptive sampling targets high residuals. Numerical experiments on traffic reconstruction confirm that enforcing causality systematically reduces the median point-wise MSE and its variability across runs, yielding improvements of nearly one order of magnitude over non-causal training in both the baseline and PD variants.
Read the full abstract ▾
Image and Video Processing · Cross-listed from cs.RO
LangGS-SLAM: Real-Time Language-Feature Gaussian Splatting SLAM
Seongbo Ha, Sibaek Lee, Kyungsu Kang et al. (+3 more)
In this paper, we propose a RGB-D SLAM system that reconstructs a
language-aligned dense feature field while sustaining low-latency tracking and mapping. First, we introduce a Top-K Rendering pipeline, a high-throughput and semantic-distortion-free method for efficiently rendering high-dimensional feature maps. To address the resulting semantic-
geometric discrepancy and mitigate the
memory consumption, we further design a multi-criteria map management strategy that prunes redundant or inconsistent Gaussians while preserving scene integrity. Finally, a hybrid field optimization framework jointly refines the
geometric and semantic fields under real-time constraints by decoupling their optimization frequencies according to field characteristics. The proposed system achieves superior
geometric fidelity compared to
geometric-only baselines and comparable semantic fidelity to offline approaches while operating at 15 FPS. Our results demonstrate that online SLAM with dense, uncompressed
language-aligned feature fields is both feasible and effective, bridging the gap between 3D perception and
language-based
reasoning.
Read the full abstract ▾
All authors Seongbo Ha, Sibaek Lee, Kyungsu Kang, Joonyeol Choi, Seungjun Tak, Hyeonwoo Yu
Signal Processing
A Pre-trained EEG-to-MEG Generative Framework for Enhancing BCI Decoding
Zhuo Li and Shuqiang Wang
Electroencephalography (
EEG) and magnetoencephalography (MEG) play important and complementary roles in non-invasive
brain-computer interface (BCI) decoding. However, compared to the low cost and portability of
EEG, MEG is more expensive and less portable, which severely limits the practical application of MEG in BCI systems. To overcome this limitation, this study proposes the first cross-modal generation framework based on
EEG-MEG spatiotemporal coupled representations to synthesize MEG signals cost-effectively. The framework first extracts general neural activity representations through a pre-trained
EEG model. Building upon these representations, the framework effectively learns the lower spatial dispersion and higher high-frequency sensitivity of MEG via the spatial focus mapping module and the broadband spectral calibration module. Experimental results demonstrate that the synthesized MEG signals show high consistency with the real MEG in both time-frequency characteristics and source space activation patterns. More importantly, downstream BCI decoding experiments demonstrate that using synthesized MEG leads to performance enhancements not only on paired
EEG-MEG
datasets but also on independent
EEG-only
datasets. Overall, this framework opens a new avenue for overcoming data bottlenecks in BCI.
Read the full abstract ▾
Image and Video Processing
Toward a Unified Semantic Loss Model for Deep JSCC-based Transmission of EO Imagery
Ti Ti Nguyen, Thanh-Dung Le, Vu Nguyen Ha et al. (+6 more)
Modern Earth Observation (EO) systems increasingly rely on high-resolution imagery to support critical applications such as environmental monitoring, disaster response, and land-use analysis. Although these applications benefit from detailed visual data, the resulting data volumes impose significant challenges on satellite communication systems constrained by limited bandwidth, power, and dynamic link conditions. To address these limitations, this paper investigates Deep Joint Source-Channel Coding (DJSCC) as an effective source-channel paradigm for the transmission of EO imagery. We focus on two complementary aspects of semantic loss in DJSCC-based systems. First, a reconstruction-centric framework is evaluated by analyzing the semantic degradation of reconstructed images under varying compression ratios and channel signal-to-noise ratios (SNR). Second, a task-oriented framework is developed by integrating DJSCC with lightweight, application-specific models (e.g., EfficientViT), with performance measured using downstream task accuracy rather than pixel-level fidelity. Based on extensive empirical analysis, we propose a unified semantic loss framework that captures both reconstruction-centric and task-oriented performance within a single model. This framework characterizes the implicit relationship between JSCC compression, channel SNR, and semantic quality, offering actionable insights for the design of robust and efficient EO imagery transmission under resource-constrained satellite links.
Read the full abstract ▾
All authors Ti Ti Nguyen, Thanh-Dung Le, Vu Nguyen Ha, Duc-Dung Tran, Hung Nguyen-Kha, Dinh-Hieu Tran, Carlos L. Marcos-Rojas, Juan C. Merlano-Duncan, Symeon Chatzinotas
Audio and Speech Processing · Cross-listed from cs.SD
PhaseCoder: Microphone Geometry-Agnostic Spatial Audio Understanding for Multimodal LLMs
Artem Dementyev, Wazeer Zulfikar, Sinan Hersek et al. (+3 more)
Current
multimodal LLMs process audio as a mono stream, ignoring the rich spatial information essential for
embodied AI. Existing spatial audio models, conversely, are constrained to fixed microphone geometries, preventing deployment across diverse devices. We present PhaseCoder, a
transformer-only spatial audio encoder that is agnostic to microphone geometry. PhaseCoder takes raw multichannel audio and microphone coordinates as inputs to perform localization and produces robust spatial
embeddings. We demonstrate that Gemma 3n LLM can be fine-tuned to reason over "Spatial Audio
Tokens" produced by PhaseCoder. We show our encoder achieves state-of-the-art results on microphone-invariant localization
benchmarks and, for the first time, enables an LLM to perform complex spatial
reasoning and targeted transcription tasks from an arbitrary microphone array.
Read the full abstract ▾
All authors Artem Dementyev, Wazeer Zulfikar, Sinan Hersek, Pascal Getreuer, Anurag Kumar, Vivek Kumar
Audio and Speech Processing
DNN-Based Online Source Counting Based on Spatial Generalized Magnitude Squared Coherence
Henri Gode and Simon Doclo
The number of active sound sources is a key parameter in many acoustic signal processing tasks, such as source localization, source separation, and multi-microphone
speech enhancement. This paper proposes a novel method for online source counting by detecting changes in the number of active sources based on spatial coherence. The proposed method exploits the fact that a single coherent source in spatially white background noise yields high spatial coherence, whereas only noise results in low spatial coherence. By applying a spatial whitening operation, the source counting problem is reformulated as a change detection task, aiming to identify the time frames when the number of active sources changes. The method leverages the generalized magnitude-squared coherence as a measure to quantify spatial coherence, providing features for a compact neural network trained to detect source count changes framewise. Simulation results with binaural hearing aids in reverberant acoustic scenes with up to 4 speakers and background noise demonstrate the effectiveness of the proposed method for online source counting.
Read the full abstract ▾
Audio and Speech Processing
Unseen but not Unknown: Using Dataset Concealment to Robustly Evaluate Speech Quality Estimation Models
Jaden Pieper and Stephen D. Voran
We introduce
Dataset Concealment (DSC), a rigorous new procedure for evaluating and interpreting objective
speech quality estimation models. DSC quantifies and decomposes the performance gap between research results and real-world application requirements, while offering context and additional insights into model behavior and
dataset characteristics. We also show the benefits of addressing the corpus effect by using the
dataset Aligner from AlignNet when training models with multiple
datasets. We demonstrate DSC and the improvements from the Aligner using nine training
datasets and nine unseen
datasets with three well-studied models: MOSNet, NISQA, and a Wav2Vec2.0-based model. DSC provides interpretable views of the generalization capabilities and limitations of models, while allowing all available data to be used at training. An additional result is that adding the 1000 parameter
dataset Aligner to the 94 million parameter Wav2Vec model during training does significantly improve the resulting model's ability to estimate
speech quality for unseen data.
Read the full abstract ▾
Systems and Control · Cross-listed from cs.LG
Safety Generalization Under Distribution Shift in Safe Reinforcement Learning: A Diabetes Testbed
Minjae Kwon, Josephine Lamp and Lu Feng
Safe Reinforcement Learning (RL)
algorithms are typically evaluated under fixed training conditions. We investigate whether training-time safety guarantees transfer to deployment under distribution shift, using diabetes management as a safety-critical testbed. We
benchmark safe RL
algorithms on a unified clinical simulator and reveal a safety generalization gap: policies satisfying constraints during training frequently violate safety requirements on unseen patients. We demonstrate that test-time shielding, which filters unsafe actions using learned dynamics models, effectively restores safety across
algorithms and patient populations. Across eight safe RL
algorithms, three diabetes types, and three age groups, shielding achieves Time-in-Range gains of 13–14% for strong baselines such as PPO-Lag and CPO while reducing clinical risk index and glucose variability. Our simulator and
benchmark provide a platform for studying safety under distribution shift in safety-critical control domains. Code is available at https://github.com/safe-autonomy-lab/GlucoSim and https://github.com/safe-autonomy-lab/GlucoAlg.
Read the full abstract ▾
Audio and Speech Processing · Cross-listed from cs.CL
Position-invariant Fine-tuning of Speech Enhancement Models with Self-supervised Speech Representations
Amit Meghanani and Thomas Hain
Integrating front-end
speech enhancement (SE) models with self-supervised learning (SSL)-based
speech models is effective for downstream tasks in noisy conditions. SE models are commonly fine-tuned using SSL representations with mean squared error (MSE) loss between enhanced and clean
speech. However, MSE is prone to exploiting positional
embeddings in SSL models, allowing the objective to be minimised through positional correlations instead of content-related information. This work frames the problem as a general limitation of self-supervised representation fine-tuning and investigates it through representation-guided SE. Two strategies are considered: (1) zero-padding, previously explored in SSL pre-training but here examined in the fine-tuning setting, and (2) speed perturbations with a soft-DTW loss. Experiments show that the soft-DTW-based approach achieves faster convergence and improved downstream performance, underscoring the importance of position-invariant fine-tuning in SSL-based
speech modelling.
Read the full abstract ▾
Image and Video Processing
Any Orthogonal Transform Will Do: Range Reduction Explains Rotation-Based Quantization for Super-Resolution
Dorsa Zeinali, Hailing Wang, Yitian Zhang et al. (+1 more)
Rotation-based quantization – multiplying weights and activations by a Hadamard
matrix before quantizing – is now standard, but why it works is explained informally, via incoherence, the central limit
theorem, or kurtosis, and rarely tested directly. We ask which property of the post-rotation distribution drives the gain, using paired hypothesis tests on the weights and activations of an image super-resolution
transformer. Three candidate properties separate cleanly. Range reduction replicates under every orthogonal transform we tried – Sylvester Hadamard, QuIP#-Paley, DCT-II, and random orthogonal
matrices – and transfers to a state-space backbone and to CNN weights. Improved normality also replicates. The increased mass near zero, by contrast, is significant only under the zero-padded Sylvester construction and disappears under unpadded transforms, so we attribute it to the padding, not the transform. The end task follows the same pattern: over a grid of three scale factors, three bitwidths and five
benchmarks, rotation improves PSNR in all 45 configurations against an otherwise identical un-rotated quantizer at the same nominal bitwidth (the rotated arm carries 0.267 bits/parameter of Sylvester padding at 4 bits and stays ahead in every
cell when charged for it), while substituting DCT-II or a random orthogonal
matrix costs only 0.06-0.07 dB, less than the padding the Hadamard is charged for. The choice of kernel is therefore an implementation decision: the Hadamard's advantage is its +/-1 entries and O(n log n) butterflies. We package the recipe as CompSRT, which improves on 2DQuant in 44 of 45 configurations at equal nominal bitwidth – charged for its padding that edge holds at 3 and 2 bits but not at 4. CondiQuant remains ahead at every operating point by 0.03-1.40 dB, which we report.
Read the full abstract ▾
All authors Dorsa Zeinali, Hailing Wang, Yitian Zhang, Yun Fu
Systems and Control
The Impact of Shared Autonomous Vehicles in Microtransit Systems: A Case Study in Atlanta
Jason Lu, Tejas Santanam, Hongzhao Guan et al. (+5 more)
Microtransit systems represent an enhancement to solve the first- and last-mile problem, integrating traditional rail and bus networks with on-demand shuttles into a flexible, integrated system. This type of demand responsive transport provides greater accessibility and higher quality levels of service compared to conventional fixed-route transit services. Advances in technology offer further opportunities to enhance microtransit performance. In particular, shared autonomous vehicles (SAVs) have the potential to transform the mobility landscape by enabling more sustainable operations, enhanced user convenience, and greater system reliability. This paper investigates the integration of SAVs in microtransit systems, advancing the technological capabilities of on-demand shuttles. A shuttle dispatching optimization model is enhanced to accommodate for driver behavior and SAV functionalities. A model predictive control approach is proposed that dynamically rebalances on-demand shuttles towards areas of higher demand without relying on vast historical data. Scenario-driven experiments are conducted using data from the MARTA Reach microtransit pilot. The results demonstrate that SAVs can elevate both service quality and user experience compared to traditional on-demand shuttles in microtransit systems.
Read the full abstract ▾
All authors Jason Lu, Tejas Santanam, Hongzhao Guan, Connor Riley, Meen-Sung Kim, Anthony Trasatti, Neda Masoud, Pascal Van Hentenryck
Systems and Control
Mean-Field Learning for Storage Aggregation
Jingguan Liu, Cong Chen, Xiaomeng Ai et al. (+3 more)
Distributed energy storage devices can be aggregated to provide operational flexibility for power systems. This requires representing a massive device population as a single, tractable surrogate that is computationally efficient and accurate. However, surrogate identification is challenging due to heterogeneity, nonconvexity, and high dimensionality of storage devices. To address these challenges, this paper develops a mean-field learning framework for storage aggregation. We interpret aggregation as the average behavior of a large storage population and show that, as the population grows, aggregate performance converges to a unique, convex mean-field limit, enabling tractable population-level modeling. This convexity further yields a price-responsive characterization of aggregate storage behavior and allows us to bound the mean-field approximation error. We construct a convex surrogate model with physically interpretable parameters that approximates the aggregate behavior of large storage populations and can be embedded directly into power system operations. Surrogate parameter identification is formulated as an optimization problem using historical price-response data, and we adopt a gradient-based
algorithm for efficient learning. Case studies validate the theoretical findings and demonstrate the effectiveness of the proposed framework in approximation accuracy and data efficiency.
Read the full abstract ▾
All authors Jingguan Liu, Cong Chen, Xiaomeng Ai, Jiakun Fang, Jinsong Wang, Jinyu Wen
Systems and Control · Cross-listed from stat.ME
Experimental Design for Matching
Chonghuan Wang
Matching mechanisms play a central role in operations management across diverse fields including education, healthcare, and online platforms. However, experimentally comparing a new matching
algorithm against a status quo presents some fundamental challenges due to matching interference, where assigning a unit in one matching may preclude its assignment in the other. In this work, we take a design-based perspective to study the design of randomized experiments to compare two predetermined matching plans on a finite population, without imposing outcome or behavioral models. We introduce the notation of a disagreement set, which captures the difference between the two matching plans, and show that it admits a unique decomposition into disjoint alternating paths and cycles with useful structural properties. Based on these properties, we propose the Alternating Path Randomized Design, which sequentially randomizes along these paths and cycles to effectively manage interference. Within a minimax framework, we optimize the conditional randomization probability and show that, for long paths, the optimal choice converges to $\sqrt{2}-1$, minimizing worst-case variance. We establish the unbiasedness of the Horvitz-Thompson
estimator and derive a finite-population Central Limit
Theorem that accommodates complex and unstable path and cycle structures as the population grows. Furthermore, we extend the design to many-to-one matchings, where capacity constraints fundamentally alter the structure of the disagreement set. Using graph-theoretic tools, including finding augmenting paths and Euler-tour decomposition on an auxiliary unbalanced directed graph, we construct feasible alternating path and cycle decompositions that allow the design and inference results to carry over.
Read the full abstract ▾
Signal Processing
Sparse Grassmannian Design for Noncoherent Codes via Schubert Cell Decomposition
Joe Asano, Yuto Hama, Hiroki Iimori et al. (+3 more)
In this paper, we propose a method for designing sparse Grassmannian codes for noncoherent multiple-input multiple-output systems. Conventional pairwise error probability formulations under uncorrelated Rayleigh fading channels fail to account for rank deficiency induced by sparse configurations. We revise these formulations to handle such cases in a unified manner. Furthermore, we derive a closed-form metric that effectively maximizes the noncoherent average mutual information (AMI) at a given signal-to-noise ratio. We focus on the fact that the Schubert
cell decomposition of the Grassmann
manifold provides a mathematically sparse property, and establish design criteria for sparse noncoherent codes based on our analyses. In numerical results, the proposed sparse noncoherent codes outperform conventional methods in terms of both symbol error rate and AMI, and asymptotically approach the performance of the optimal Grassmannian constellations in the high-signal-to-noise ratio regime. Moreover, they reduce the time and space complexity, which does not scale with the number of transmit
antennas.
Read the full abstract ▾
All authors Joe Asano, Yuto Hama, Hiroki Iimori, Chandan Pradhan, Szabolcs Malomsoky, Naoki Ishikawa
Audio and Speech Processing · Cross-listed from cs.CL
asr_eval: Algorithms and tools for multi-reference and streaming speech recognition evaluation
Oleg Sedukhin and Andrey Kostin
We propose several improvements to the
speech recognition evaluation. First, we propose a string
alignment algorithm that supports both multi-reference labeling, arbitrary-length insertions and better word
alignment. This is especially useful for non-Latin
languages, those with rich word formation, to label cluttered or longform
speech. Secondly, we collect a novel test set DiverseSpeech-Ru of longform in-the-wild Russian
speech with careful multi-reference labeling. We also perform multi-reference relabeling of popular Russian tests set and study fine-tuning dynamics on its corresponding train set. We demonstrate that the model often adopts to
dataset-specific labeling, causing an illusion of metric improvement. Based on the improved word
alignment, we develop tools to evaluate streaming
speech recognition and to align multiple transcriptions to compare them visually. Additionally, we provide uniform wrappers for many offline and streaming
speech recognition models. Our code will be made publicly available.
Read the full abstract ▾
Image and Video Processing
Denoising and Baseline Correction of Low-Scan FTIR Spectra: A Benchmark of Deep Learning Models Against Traditional Signal Processing
Azadeh Mokari, Shravan Raghunathan, Artem Shydliukh et al. (+3 more)
High-quality
Fourier Transform Infrared (FTIR) imaging usually needs extensive signal averaging to reduce noise and drift which severely limits clinical speed. Deep learning can accelerate imaging by reconstructing spectra from rapid, single-scan inputs. However, separating noise and baseline drift simultaneously without ground truth is an ill-posed inverse problem. Standard black-box architectures often rely on statistical approximations that introduce spectral
hallucinations or fail to generalize to unstable atmospheric conditions. To solve these issues we propose a physics-informed cascade Unet that separates denoising and baseline correction tasks using a new, deterministic Physics Bridge. This architecture forces the network to separate random noise from chemical signals using an embedded SNIP layer to enforce spectroscopic constraints instead of learning statistical approximations. We benchmarked this approach against a standard single Unet and a traditional Savitzky-Golay/SNIP workflow. We used a
dataset of human hypopharyngeal carcinoma
cells (FaDu). The cascade model outperformed all other methods, achieving a 51.3% reduction in RMSE compared to raw single-scan inputs, surpassing both the single Unet (40.2%) and the traditional workflow (33.7%). Peak-aware metrics show that the cascade architecture eliminates spectral
hallucinations found in standard deep learning. It also preserves peak intensity with much higher fidelity than traditional smoothing. These results show that the cascade Unet is a robust solution for diagnostic-grade FTIR imaging. It enables imaging speeds 32 times faster than current methods.
Read the full abstract ▾
All authors Azadeh Mokari, Shravan Raghunathan, Artem Shydliukh, Oleg Ryabchykov, Christoph Krafft, Thomas Bocklitz
Image and Video Processing
ECGFlowCMR: Pretraining with ECG-Generated Cine CMR Helps Cardiac Disease Classification and Phenotype Prediction
Xiaocheng Fang, Zhengyao Ding, Guangkun Nie et al. (+9 more)
Cardiac
Magnetic Resonance (CMR) imaging provides a comprehensive assessment of cardiac structure and function but remains constrained by high acquisition costs and reliance on expert annotations, limiting the availability of large-scale labeled
datasets. In contrast, electrocardiograms (ECGs) are inexpensive, widely accessible, and offer a promising modality for conditioning the generative synthesis of cine CMR. To this end, we propose ECGFlowCMR, a novel ECG-to-CMR generative framework that integrates a Phase-Aware Masked Autoencoder (PA-MAE) and an Anatomy-Motion Disentangled Flow (AMDF) to address two fundamental challenges: (1) the cross-modal temporal mismatch between multi-beat ECG recordings and single-cycle CMR sequences, and (2) the anatomical observability gap due to the limited structural information inherent in ECGs. Extensive experiments on the UK Biobank and a proprietary clinical
dataset demonstrate that ECGFlowCMR can generate realistic cine CMR sequences from ECG inputs, enabling scalable pretraining and improving performance on downstream cardiac disease classification and phenotype prediction tasks.
Read the full abstract ▾
All authors Xiaocheng Fang, Zhengyao Ding, Guangkun Nie, Jieyi Cai, Yujie Xiao, Bo Liu, Jiarui Jin, Haoyu Wang, Shun Huang, Ting Chen, Hongyan Li, Shenda Hong
Audio and Speech Processing · Cross-listed from cs.SD
Text-only adaptation in LLM-based ASR through text denoising
Andrés Carofilis, Sergio Burdisso, Esaú Villatoro-Tello et al. (+8 more)
Adapting large
language model (LLM)-based automatic
speech recognition (ASR) systems to new domains using text-only data is a significant yet underexplored challenge. Standard fine-tuning of the LLM on the target domain text often disrupts the critical
alignment between the
speech and text modality learned by the projector, degrading performance. We introduce a novel text-only adaptation method that frames this process as a text denoising task. Our approach trains the LLM to recover clean transcripts from noisy inputs. This process effectively adapts the model to a target domain while preserving cross-modal
alignment. Our solution is lightweight, requiring no architectural changes or additional parameters. Extensive evaluation on two
datasets demonstrates up to 22.1% relative improvement, outperforming recent state-of-the-art text-only adaptation methods.
Read the full abstract ▾
All authors Andrés Carofilis, Sergio Burdisso, Esaú Villatoro-Tello, Shashi Kumar, Kadri Hacioglu, Srikanth Madikeri, Pradeep Rangappa, Manjunath K E, Petr Motlicek, Shankar Venkatesan, Andreas Stolcke
Audio and Speech Processing
Reducing Prompt Sensitivity in LLM-based Speech Recognition Through Learnable Projection
Sergio Burdisso, Esaú Villatoro-Tello, Shashi Kumar et al. (+7 more)
LLM-based automatic
speech recognition (ASR), a well-established approach, connects
speech foundation models to large
language models (LLMs) through a
speech-to-LLM projector, yielding promising results. A common design choice in these architectures is the use of a fixed, manually defined
prompt during both training and inference. This setup not only enables applicability across a range of practical scenarios, but also helps maximize model performance. However, the impact of
prompt design remains underexplored. This paper presents a comprehensive analysis of commonly used
prompts across diverse
datasets, showing that
prompt choice significantly affects ASR performance and introduces instability, with no single
prompt performing best across all cases. Inspired by the
speech-to-LLM projector, we propose a
prompt projector module, a simple, model-agnostic extension that learns to project
prompt embeddings to more effective regions of the LLM input space, without modifying the underlying LLM-based ASR model. Experiments on four
datasets show that the addition of a
prompt projector consistently improves performance, reduces variability, and outperforms the best manually selected
prompts.
Read the full abstract ▾
All authors Sergio Burdisso, Esaú Villatoro-Tello, Shashi Kumar, Srikanth Madikeri, Andrés Carofilis, Pradeep Rangappa, Manjunath K E, Kadri Hacioglu, Petr Motlicek, Andreas Stolcke
Audio and Speech Processing · Cross-listed from cs.SD
A Study of Data Selection Strategies for Pre-training Self-Supervised Speech Models
Ryan Whetten, Titouan Parcollet, Marco Dinarelli et al. (+1 more)
Self-supervised learning (SSL) has transformed
speech processing, yet its reliance on massive pre-training
datasets remains a bottleneck. While robustness is often attributed to scale and diversity, the role of the data distribution is less understood. We systematically examine how curated subsets of pre-training data influence Automatic
Speech Recognition (ASR) performance. Surprisingly, optimizing for acoustic, speaker, or linguistic diversity yields no clear improvements over random sampling. Instead, we find that prioritizing the longest utterances achieves superior ASR results while using only half the original
dataset, reducing pre-training time by 24% on a large corpora. These findings suggest that for pre-training
speech SSL models, data length is a more critical factor than either data diversity or overall data quantity for performance and efficiency, offering a new perspective for data selection strategies in SSL
speech processing.
Read the full abstract ▾
All authors Ryan Whetten, Titouan Parcollet, Marco Dinarelli, Yannick Estève
Audio and Speech Processing · Cross-listed from cs.SD
SW-ASR: A Context-Aware Hybrid ASR Pipeline for Robust Single Word Speech Recognition
Manali Sharma, Riya Naik and Buvaneshwari G
Single-word Automatic
Speech Recognition (ASR) is a challenging task due to the lack of linguistic context and sensitivity to noise, pronunciation variation, and channel artifacts, especially in low-resource, communication-critical domains such as healthcare and emergency response. This paper reviews recent deep learning approaches and proposes a modular framework for robust single-word detection. The system combines denoising and normalization with a hybrid ASR front end (Whisper + Vosk) and a verification layer designed to handle out-of-vocabulary words and degraded audio. The verification layer supports multiple matching strategies, including
embedding similarity, edit distance, and LLM-based matching with optional contextual guidance. We evaluate the framework on the Google
Speech Commands
dataset and a curated real-world
dataset collected from telephony and messaging platforms under bandwidth-limited conditions. Results show that while the hybrid ASR front end performs well on clean audio, the verification layer significantly improves accuracy on noisy and compressed channels. Context-guided and LLM-based matching yield the largest gains, demonstrating that lightweight verification and context mechanisms can substantially improve single-word ASR robustness without sacrificing latency required for real-time telephony applications.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.LG
PatchFormer: A Patch-Based Time Series Foundation Model with Hierarchical Masked Reconstruction and Cross-Domain Transfer Learning for Zero-Shot Multi-Horizon Forecasting
Olaf Yunus Laitinen Imanov, Derya Umut Kulali and Taner Yilmaz
Time series
forecasting is a fundamental problem with applications in
climate, energy, healthcare, and finance. Many existing approaches require domain-specific feature engineering and substantial labeled data for each task. We introduce PatchFormer, a patch-based time series foundation model that uses hierarchical masked reconstruction for self-supervised pretraining and lightweight adapters for efficient transfer. PatchFormer segments time series into patches and learns multiscale temporal representations with learnable aggregation across temporal scales. Pretraining uses masked patch reconstruction with dynamic masking and objectives that encourage both local accuracy and global consistency, followed by cross-domain knowledge
distillation. Experiments on 24
benchmark datasets spanning weather, energy, traffic, finance, and healthcare demonstrate state-of-the-art zero-shot multi-horizon
forecasting, reducing mean squared error by 27.3 percent relative to strong baselines while requiring 94 percent less task-specific training data. The model exhibits near log-linear scaling with more pretraining data up to 100 billion points and processes length-512 sequences 3.8x faster than full-sequence
transformers.
Read the full abstract ▾
Signal Processing
Statistical Analysis of the Extensive Cancellation Algorithm for Passive Radar Using an Imperfect Reference Signal
Mats Viberg, Daniele Gerosa, Tomas McKelvey et al. (+1 more)
Passive radar systems have received tremendous
attention over the past few decades, due to their low cost and ability to remain covert during operation. Such systems rely on a so-called Illuminator-of-Opportunity (IO), for example, a commercial TV station. We consider a network of Receiving Nodes (RN) without spatial resolution capability, which receives the direct signal and reflections from both stationary objects (clutter) and possible targets. After suitable preprocessing, the RNs transmit information to a Fusion Center (FC) that performs the final target detection, localization and tracking. Several methods for target localization have been proposed in the literature, and our focus is on the seminal Extensive Cancellation
Algorithm (ECA). In this approach, each RN collects information about target parameters, while canceling interference using a projection. This is done by exploiting a separate Reference Channel (RC), which captures the IO signal without interference apart from receiver noise. We derive the statistical properties of the ECA parameter estimates under the assumption of a high Signal-to-Noise Ratio (SNR), and we give a sufficient condition for the SNR in the RC to enable statistically efficient estimates. The theoretical results are corroborated through computer simulations, which indicate that the theory agrees well with empirical results under practical operating conditions. The contributions of this paper can be used, for example, to design experimental setups for feasibility studies and to inform system design for achieving a desired localization accuracy.
Read the full abstract ▾
All authors Mats Viberg, Daniele Gerosa, Tomas McKelvey, Thomas Eriksson
Signal Processing
AI-Driven Design of Stacked Intelligent Metasurfaces for Software-Defined Radio Applications
Ivan Iudice, Giacinto Gelli and Donatella Darsena
The integration of reconfigurable intelligent surfaces (RIS) into future wireless communication systems offers promising capabilities in dynamic environment shaping and spectrum efficiency. In this work, we present a consistent implementation of a stacked intelligent metasurface (SIM) model within the NVIDIA's AI-native framework Sionna for 6G physical layer research. Our implementation allows simulation and learning-based optimization of SIM-assisted communication channels in fully differentiable and GPU-accelerated environments, enabling end-to-end training for cognitive and software-defined
radio (SDR) applications. We describe the architecture of the SIM model, including its integration into the TensorFlow-based pipeline, and showcase its use in closed-loop learning scenarios involving adaptive beamforming and dynamic reconfiguration. Benchmarking results are provided for various deployment scenarios, highlighting the model's effectiveness in enabling intelligent control and signal enhancement in non-terrestrial-network (NTN) propagation environments. This work demonstrates a scalable, modular approach for incorporating intelligent metasurfaces into modern AI-accelerated SDR systems and paves the way for future hardware-in-the-loop experiments.
Read the full abstract ▾