Peer Revenue Fully funded

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Electrical Engineering and Systems Science

eess Watch Electrical Engineering and Systems Science

Wednesday 25 February

Systems and Control · Cross-listed from cs.RO

LiLo-VLA: Compositional Long-Horizon Manipulation via Linked Object-Centric Policies

Yue Yang, Shuo Cheng, Yu Fang et al. (+4 more)

General-purpose robots must master long-horizon manipulation, defined as tasks involving multiple kinematic structure changes (e.g., attaching or detaching objects) in unstructured environments. While Vision-Language-Action (VLA) models offer the potential to master diverse atomic skills, they struggle with the combinatorial complexity of sequencing them and are prone to cascading failures due to environmental sensitivity. To address these challenges, we propose LiLo-VLA (Linked Local VLA), a modular framework capable of zero-shot generalization to novel long-horizon tasks without ever being trained on them. Our approach decouples transport from interaction: a Reaching Module handles global motion, while an Interaction Module employs an object-centric VLA to process isolated objects of interest, ensuring robustness against irrelevant visual features and invariance to spatial configurations. Crucially, this modularity facilitates robust failure recovery through dynamic replanning and skill reuse, effectively mitigating the cascading errors common in end-to-end approaches. We introduce a 21-task simulation benchmark consisting of two challenging suites: LIBERO-Long++ and Ultra-Long. In these simulations, LiLo-VLA achieves a 69% average success rate, outperforming Pi0.5 by 41% and OpenVLA-OFT by 67%. Furthermore, real-world evaluations across 8 long-horizon tasks demonstrate an average success rate of 85%. Project page: https://yy-gx.github.io/LiLo-VLA/.
Read the full abstract

All authors Yue Yang, Shuo Cheng, Yu Fang, Homanga Bharadhwaj, Mingyu Ding, Gedas Bertasius, Daniel Szafir

Systems and Control

Optimal Real-Time Fusion of Time-Series Data Under Rényi Differential Privacy

Chuanghong Weng and Ehsan Nekouei

In this paper, we investigate the optimal real-time fusion of data collected by multiple sensors. In our set-up, the sensor measurements are considered to be private and are jointly correlated with an underlying process. A fusion center combines the private sensor measurements and releases its output to an honest-but-curious party, which is responsible for estimating the state of the underlying process based on the fusion center's output. The privacy leakage incurred by the fusion policy is quantified using Rényi differential privacy. We formulate the privacy-aware fusion design as a constrained finite-horizon optimization problem, in which the fusion policy and the state estimation are jointly optimized to minimize the state estimation error subject to a total privacy budget constraint. We derive the constrained optimality conditions for the proposed optimization problem and use them to characterize the structural properties of the optimal fusion policy. Unlike classical differential privacy mechanisms, the optimal fusion policy is shown to adaptively allocates the privacy budget and regulates the adversary's belief in a closed-loop manner. To reduce the computational burden of solving the resulting constrained optimality equations, we parameterize the fusion policy using a structured Gaussian distribution and show that the parameterized fusion policy satisfies the privacy constraint. We further develop a numerical algorithm to jointly optimize the fusion policy and state estimator. Finally, we demonstrate the effectiveness of the proposed fusion framework through a traffic density estimation case study.
Read the full abstract

Image and Video Processing

Deep Unfolding Real-Time Super-Resolution Using Subpixel-Shift Twin Image and Convex Self-Similarity Prior

Chia-Hsiang Lin, Wei-Chih Liu, Yu-En Chiu et al. (+1 more)

Multi-image super-resolution (MISR) is a critical technique for satellite remote sensing. In the perspective of information, twin-image super-resolution (TISR) is regarded as the most challenging MISR scenario, having crucial applications like the SPOT-5 supermode imaging. In TISR, an image is super-resolved by its subpixel-shift counterpart (i.e., twin image), where the two images are typically offset by half a pixel both horizontally and vertically. We formulate the less investigated TISR using a convex criterion, which is implemented using a novel deep unfolding network. In the unfolding, an embedded simple shift operator trickily addresses the coupled TISR data-fitting terms, and a transformer trained with a convex self-similarity loss function elegantly implements the proximal mapping induced by the TISR regularizer. The proposed convex self-similarity unfolding supermode super-resolution (COSUP) algorithm is interpretable and achieves state-of-the-art performance with very fast millisecond-level computational time. COSUP is also tested on real-world data, for which the subpixel shifts would not be spatially uniform, with results showing great superiority over the official CNES supermode imaging product in terms of credible metrics (e.g., natural image quality evaluator, NIQE). Source codes: https://github.com/IHCLab/COSUP.
Read the full abstract

All authors Chia-Hsiang Lin, Wei-Chih Liu, Yu-En Chiu, Jhao-Ting Lin

Image and Video Processing

Perceptual Quality Optimization of Image Super-Resolution

Wei Zhou, Yixiao Li, Hadi Amirpour et al. (+4 more)

Single-image super-resolution (SR) has achieved remarkable progress with deep learning, yet most approaches rely on distortion-oriented losses or heuristic perceptual priors, which often lead to a trade-off between fidelity and visual quality. To address this issue, we propose an Efficient Perceptual Bi-directional Attention Network (Efficient-PBAN) that explicitly optimizes SR towards human-preferred quality. Unlike patch-based quality models, Efficient-PBAN avoids extensive patch sampling and enables efficient image-level perception. The proposed framework is trained on our self-constructed SR quality dataset that covers a wide range of state-of-the-art SR methods with corresponding human opinion scores. Using this dataset, Efficient-PBAN learns to predict perceptual quality in a way that correlates strongly with subjective judgments. The learned metric is further integrated into SR training as a differentiable perceptual loss, enabling closed-loop alignment between reconstruction and perceptual assessment. Extensive experiments demonstrate that our approach delivers superior perceptual quality. Code is publicly available at https://github.com/Lighting-YXLI/Efficient-PBAN.
Read the full abstract

All authors Wei Zhou, Yixiao Li, Hadi Amirpour, Xiaoshuai Hao, Jiang Liu, Peng Wang, Hantao Liu

Audio and Speech Processing

A Knowledge-Driven Approach to Music Segmentation, Music Source Separation and Cinematic Audio Source Separation

Chun-wei Ho, Sabato Marco Siniscalchi, Kai Li et al. (+1 more)

We propose a knowledge-driven, model-based approach to segmenting audio into single-category and mixed-category chunks with applications to source separation. "Knowledge" here denotes information associated with the data, such as music scores. "Model" here refers to tool that can be used for audio segmentation and recognition, such as hidden Markov models. In contrast to conventional learning that often relies on annotated data with given segment categories and their corresponding boundaries to guide the learning process, the proposed framework does not depend on any pre-segmented training data and learns directly from the input audio and its related knowledge sources to build all necessary models autonomously. Evaluation on simulation data shows that score-guided learning achieves very good music segmentation and separation results. Tested on movie track data for cinematic audio source separation also shows that utilizing sound category knowledge achieves better separation results than those obtained with data-driven techniques without using such information.
Read the full abstract

All authors Chun-wei Ho, Sabato Marco Siniscalchi, Kai Li, Chin-Hui Lee

Audio and Speech Processing

iMiGUE-Speech: A Spontaneous Speech Dataset for Affective Analysis

Sofoklis Kakouros, Fang Kang and Haoyu Chen

This work presents iMiGUE-Speech, an extension of the iMiGUE dataset that provides a spontaneous affective corpus for studying emotional and affective states. The new release focuses on speech and enriches the original dataset with additional metadata, including speech transcripts, speaker-role separation between interviewer and interviewee, and word-level forced alignments. Unlike existing emotional speech datasets that rely on acted or laboratory-elicited emotions, iMiGUE-Speech captures spontaneous affect arising naturally from real match outcomes. To demonstrate the utility of the dataset and establish initial benchmarks, we introduce two evaluation tasks for comparative assessment: speech emotion recognition and transcript-based sentiment analysis. These tasks leverage state-of-the-art pre-trained representations to assess the dataset's ability to capture spontaneous affective states from both acoustic and linguistic modalities. iMiGUE-Speech can also be synchronously paired with micro-gesture annotations from the original iMiGUE dataset, forming a uniquely multimodal resource for studying speech-gesture affective dynamics. The extended dataset is available at https://github.com/CV-AC/imigue-speech.
Read the full abstract

Systems and Control · Cross-listed from cs.RO

Vector Fields for Path Following on Lie Groups with Application in Robot Control

Felipe Bartelt, Luciano C. A. Pimenta, Weijia Yao et al. (+1 more)

Many robotic systems allow independent control of position and orientation (pose), including omnidirectional aerial vehicles, underwater robots, and manipulator end-effectors. In many applications, these systems must follow a continuous sequence of poses, leading to either trajectory-tracking or path following formulations. Compared to trajectory-tracking, path following offers important practical advantages. In particular, we focus on the problem of path following on Lie groups. Considering the robots as rigid bodies moving in the 3D space, this path-following problem can be posed as a problem of designing guiding vector fields on the matrix Lie group SE(3). In this paper, we develop a general vector-field framework for path following on connected matrix Lie groups, of which SE(3) is a prominent special case. The proposed vector field guarantees convergence to a desired parametric curve from almost all initial conditions while ensuring continuous motion along the path. Furthermore, another interesting feature is that, as opposed to previous works, the control input is "minimal" in terms of representation and closer to the engineering application (e.g., the body twist in the case SE(3)). After establishing the general case, the framework is then specialized to SE(3), of special interest in robotics, yielding an efficient algorithm suitable for real-time robotic control. Experiments with a robotic manipulator tracking complex pose paths demonstrate the effectiveness of the approach. An open-source implementation is also provided.
Read the full abstract

All authors Felipe Bartelt, Luciano C. A. Pimenta, Weijia Yao, Vinicius M. Gonçalves

Signal Processing

Semi-Gridless Variational Bayes Channel Estimation in XL-MIMO: Near-Field Modeling and Inference

Van-Chung Luu, Toan-Van Nguyen, Nuria González-Prelcic et al. (+1 more)

Extremely large antenna arrays and high-frequency operation are two key technologies that advance performance metrics such as higher data rates, lower latency, and wider coverage in sixth-generation communications. However, the adoption of these technologies fundamentally changes the characteristics of wavefronts, forcing communication systems to operate in the near-field region. The transition from planar far-field communications to spherical near-field propagation necessitates novel channel estimation algorithms to fully exploit the unique features of spherical wavefronts for advanced transceiver design. To this end, we propose a novel semi-gridless channel estimation approach based on a variational Bayesian (VB) inference framework. Specifically, we reformulate the near-field channel model for both uniform linear arrays and uniform planar arrays into separate direction-of-arrival (DoAs) and distance components. Building on these new representations, we employ a gridless approach for DoAs estimation using a von Mises distribution, and a coarse-to-fine grid search for distance estimation. We then develop a semi-gridless variational Bayesian (SG-VB) algorithm with efficient update rules that enables accurate channel reconstruction. Simulation results validate the effectiveness of the proposed SG-VB algorithm, demonstrating enhanced near-field channel reconstruction accuracy and superior estimation performance for both DoAs and distance components embedded in near-field channels.
Read the full abstract

All authors Van-Chung Luu, Toan-Van Nguyen, Nuria González-Prelcic, Duy H. N. Nguyen

Tuesday 24 February

Signal Processing

NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning

Guoan Wang, Shihao Yang, Jun-en Ding et al. (+2 more)

Electroencephalography (EEG) provides a non-invasive window into neural dynamics at high temporal resolution and plays a pivotal role in clinical neuroscience research. Despite this potential, prevailing computational approaches to EEG analysis remain largely confined to task-specific classification objectives or coarse-grained pattern recognition, offering limited support for clinically meaningful interpretation. To address these limitations, we introduce NeuroNarrator, the first generalist EEG-to-text foundation model designed to translate electrophysiological segments into precise clinical narratives. A cornerstone of this framework is the curation of NeuroCorpus-160K, the first harmonized large-scale resource pairing over 160,000 EEG segments with structured, clinically grounded natural-language descriptions. Our architecture first aligns temporal EEG waveforms with spatial topographic maps via a rigorous contrastive objective, establishing spectro-spatially grounded representations. Building on this grounding, we condition a Large Language Model through a state-space-inspired formulation that integrates historical temporal and spectral context to support coherent clinical narrative generation. This approach establishes a principled bridge between continuous signal dynamics and discrete clinical language, enabling interpretable narrative generation that facilitates expert interpretation and supports clinical reporting workflows. Extensive evaluations across diverse benchmarks and zero-shot transfer tasks highlight NeuroNarrator's capacity to integrate temporal, spectral, and spatial dynamics, positioning it as a foundational framework for time-frequency-aware, open-ended clinical interpretation of electrophysiological data.
Read the full abstract

All authors Guoan Wang, Shihao Yang, Jun-en Ding, Hao Zhu, Feng Liu

Systems and Control

PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow with Continual Learning

Chidozie Ezeakunne, Jose E. Tabarez, Reeju Pokharel et al. (+1 more)

Solving the alternating current power flow equations in real time is essential for secure grid operation, yet classical Newton-Raphson solvers can be slow under stressed conditions. Existing graph neural networks for power flow are typically trained on a single system and often degrade on different systems. We present PowerModelsGAT-AI, a physics-informed graph attention network that predicts bus voltages and generator injections. The model uses bus-type-aware masking to handle different bus types and balances multiple loss terms, including a power-mismatch penalty, using learned weights. We evaluate the model on 14 benchmark systems (4 to 6,470 buses) and train a unified model on 13 of these under N-2 (two-branch outage) conditions, achieving an average normalized mean absolute error of 0.89% for voltage magnitudes and R^2 > 0.99 for voltage angles. We also show continual learning: when adapting a base model to a new 1,354-bus system, standard fine-tuning causes severe forgetting with error increases exceeding 1000% on base systems, while our experience replay and elastic weight consolidation strategy keeps error increases below 2% and in some cases improves base-system performance. Interpretability analysis shows that learned attention weights correlate with physical branch parameters (susceptance: r = 0.38; thermal limits: r = 0.22), and feature importance analysis supports that the model captures established power flow relationships.
Read the full abstract

All authors Chidozie Ezeakunne, Jose E. Tabarez, Reeju Pokharel, Anup Pandey

Systems and Control · Cross-listed from cs.DC

Autonomous Edge-Deployed AI Agents for Electric Vehicle Charging Infrastructure Management

Mohammed Cherifi

Public EV charging infrastructure suffers from significant failure rates – with field studies reporting up to 27.5% of DC fast chargers non-functional – and multi-day mean time to resolution, imposing billions in annual economic burden. Cloud-centric architectures cannot achieve the latency, reliability, and bandwidth characteristics required for autonomous operation. We present Auralink SDC (Software-Defined Charging), an architecture deploying domain-specialized AI agents at the network edge for autonomous charging infrastructure management. Key contributions include: (1) Confidence-Calibrated Autonomous Resolution (CCAR), enabling autonomous remediation with formal false-positive bounds; (2) Adaptive Retrieval-Augmented Reasoning (ARA), combining dense and sparse retrieval with dynamic context allocation; (3) Auralink Edge Runtime, achieving sub-50ms TTFT on commodity hardware under PREEMPT_RT constraints; and (4) Hierarchical Multi-Agent Orchestration (HMAO). Implementation uses AuralinkLM models fine-tuned via QLoRA on a domain corpus spanning OCPP 1.6/2.0.1, ISO 15118, and operational incident histories. Evaluation on 18,000 labeled incidents in a controlled environment establishes 78% autonomous incident resolution, 87.6% diagnostic accuracy, and 28-48ms TTFT latency (P50). This work presents architecture and implementation patterns for edge-deployed industrial AI systems with safety-critical constraints.
Read the full abstract

Audio and Speech Processing

LMU-Based Sequential Learning and Posterior Ensemble Fusion for Cross-Domain Infant Cry Classification

Niloofar Jazaeri, Hilmi R. Dajani, Marco Janeczek et al. (+1 more)

Decoding infant cry causes remains challenging for healthcare monitoring due to short nonstationary signals, limited annotations, and strong domain shifts across infants and datasets. We propose a compact acoustic framework that fuses mel-frequency cepstral coefficients (MFCCs), short-time Fourier transform (STFT) features, and fundamental-frequency (F0) contours within a multi-branch convolutional neural network (CNN) encoder, and models temporal dynamics using an enhanced Legendre Memory Unit (LMU). Compared to LSTMs, the LMU backbone provides stable sequence modeling with substantially fewer recurrent parameters, supporting efficient deployment. To improve cross-dataset generalization, we introduce calibrated posterior ensemble fusion with entropy-gated weighting to preserve domain-specific expertise while mitigating dataset bias. Experiments on Baby2020 and Baby Crying demonstrate improved macro-F1 under cross-domain evaluation, along with leakage aware splits and real-time feasibility for on-device monitoring.
Read the full abstract

All authors Niloofar Jazaeri, Hilmi R. Dajani, Marco Janeczek, Martin Bouchard

Image and Video Processing · Cross-listed from cs.CV

Leveraging GenAI for Segmenting and Labeling Centuries-old Technical Documents

Carlos Monroy and Benjamin Navarro

Image segmentation and image recognition are well established computational techniques in the broader discipline of image processing. Segmentation allows to locate areas in an image, while recognition identifies specific objects within an image. These techniques have shown remarkable accuracy with modern images, mainly because the amount of training data is vast. Achieving similar accuracy in digitized images of centuries-old documents is more challenging. This difficulty is due to two main reasons: first, the lack of sufficient training data, and second, because the degree of specialization in a given domain. Despite these limitations, the ability to segment and recognize objects in these collections is important for automating the curation, cataloging, and dissemination of knowledge, making the contents of priceless collections accessible to scholars and the general public. In this paper, we report on our ongoing work in segmenting and labeling images pertaining to shipbuilding treatises from the XVI and XVII centuries, a historical period known as the Age of Exploration. To this end, we leverage SAM2 for image segmentation; Florence2 and ChatGPT for labeling; and a specialized ontology ontoShip and glossary glosShip of nautical architecture for enhancing the labeling process. Preliminary results demonstrate the potential of marrying these technologies for improving curation and retrieval of priceless historical documents. We also discuss the challenges and limitations encountered in this approach and ideas on how to overcome them in the future.
Read the full abstract

Image and Video Processing · Cross-listed from cs.CV

From Scale to Speed: Adaptive Test-Time Scaling for Image Editing

Xiangyan Qu, Zhenlong Yuan, Jing Tang et al. (+9 more)

Image Chain-of-Thought (Image-CoT) is a test-time scaling paradigm that improves image generation by extending inference time. Most Image-CoT methods focus on text-to-image (T2I) generation. Unlike T2I generation, image editing is goal-directed: the solution space is constrained by the source image and instruction. This mismatch causes three challenges when applying Image-CoT to editing: inefficient resource allocation with fixed sampling budgets, unreliable early-stage verification using general MLLM scores, and redundant edited results from large-scale sampling. To address this, we propose ADaptive Edit-CoT (ADE-CoT), an on-demand test-time scaling framework to enhance editing efficiency and performance. It incorporates three key strategies: (1) a difficulty-aware resource allocation that assigns dynamic budgets based on estimated edit difficulty; (2) edit-specific verification in early pruning that uses region localization and caption consistency to select promising candidates; and (3) depth-first opportunistic stopping, guided by an instance-specific verifier, that terminates when intent-aligned results are found. Extensive experiments on three SOTA editing models (Step1X-Edit, BAGEL, FLUX.1 Kontext) across three benchmarks show that ADE-CoT achieves superior performance-efficiency trade-offs. With comparable sampling budgets, ADE-CoT obtains better performance with more than 2x speedup over Best-of-N.
Read the full abstract

All authors Xiangyan Qu, Zhenlong Yuan, Jing Tang, Rui Chen, Datao Tang, Meng Yu, Lei Sun, Yancheng Bai, Xiangxiang Chu, Gaopeng Gou, Gang Xiong, Yujun Cai

Systems and Control · Cross-listed from cs.LG

Improving Spatial Allocation for Energy System Coupling with Graph Neural Networks

Xuanhao Mu, Jakob Geiges, Nan Liu et al. (+2 more)

In energy system analysis, coupling models with mismatched spatial resolutions is a significant challenge. A common solution is assigning weights to high-resolution geographic units for aggregation, but traditional models are limited by using only a single geospatial attribute. This paper presents an innovative method employing a self-supervised Heterogeneous Graph Neural Network to address this issue. This method models high-resolution geographic units as graph nodes, integrating various geographical features to generate physically meaningful weights for each grid point. These weights enhance the conventional Voronoi-based allocation method, allowing it to go beyond simply geographic proximity by incorporating essential geographic information.In addition, the self-supervised learning paradigm overcomes the lack of accurate ground-truth data. Experimental results demonstrate that applying weights generated by this method to cluster-based Voronoi Diagrams significantly enhances scalability, accuracy, and physical plausibility, while increasing precision compared to traditional methods.
Read the full abstract

All authors Xuanhao Mu, Jakob Geiges, Nan Liu, Thorsten Schlachter, Veit Hagenmeyer

Signal Processing · Cross-listed from physics.ins-det

Machine Learning on Heterogeneous, Edge, and Quantum Hardware for Particle Physics (ML-HEQUPP)

Julia Gonski, Jennifer Ott, Shiva Abbaszadeh et al. (+118 more)

The next generation of particle physics experiments will face a new era of challenges in data acquisition, due to unprecedented data rates and volumes along with extreme environments and operational constraints. Harnessing this data for scientific discovery demands real-time inference and decision-making, intelligent data reduction, and efficient processing architectures beyond current capabilities. Crucial to the success of this experimental paradigm are several emerging technologies, such as artificial intelligence and machine learning (AI/ML), silicon microelectronics, and the advent of quantum algorithms and processing. Their intersection includes areas of research such as low-power and low-latency devices for edge computing, heterogeneous accelerator systems, reconfigurable hardware, novel codesign and synthesis strategies, readout for cryogenic or high-radiation environments, and analog computing. This white paper presents a community-driven vision to identify and prioritize research and development opportunities in hardware-based ML systems and corresponding physics applications, contributing towards a successful transition to the new data frontier of fundamental science.
Read the full abstract

All authors Julia Gonski, Jennifer Ott, Shiva Abbaszadeh, Sagar Addepalli, Matteo Cremonesi, Jennet Dickinson, Giuseppe Di Guglielmo, Yigit Erdem Ertorer, Lindsey Gray, Ryan Herbst, Christian Herwig, Tae Min Hong, Benedikt Maier, Maryam Bayat Makou, David W. Miller, Mark S. Neubauer, Cristian Pena, Dylan Rankin, Seon-Hee Seo, Giordon Stark, Alexander Tapper, Audrey Corbeil Therrien, Ioannis Xiotidis, Keisuke Yoshihara, Nural Akchurin, Carlos Arguelles, Saptaparna Bhattacharya, Lorenzo Borella, Christian Boutan, Tom Braine, James Brau, Marco Carminati, Antonio Chahine, Yuan-Tang Chou, Talal Ahmed Chowdhury, Seokju Chung, Ryan Coffee, Alberto Coppi, Mariarosaria D'Alfonso, Abhilasha Dave, Chance Desmet, Angela Di Fulvio, Karri DiPetrillo, Javier Duarte, Auralee Edelen, Jan Eysermans, Yongbin Feng, Emmett Forrestel, Julian Garcia Pardinas, Loredana Gastaldo, Lino Gerlach, Abarajithan Gnaneswaran, Loukas Gouskos, Katya Govorkova, Carl Grace, Christopher Grant, Philip Harris, Ciaran Hasnip, Timon Heim, Abraham Holtermann, Gian Michele Innocenti, Koji Ishidoshiro, Miaochen Jin, Jyothisraj Johnson, Stephen Jones, Andreas Jung, Nicholas Kamp, Georgia Karagiorgi, Ryan Kastner, Doojin Kim, Kyoungchul Kong, Katie Kudela, Bo-Cheng Lai, Yun-Tsung Lai, Jelena Lalic, Tommy Lam, Jeffrey Lazar, Aobo Li, Zepeng Li, Haoyun Liu, Qibin Liu, Vladimir Loncar, Zhenghua Ma, Luca Macchiarulo, Christopher Madrid, Prashansa Mukim, Victoria Nguyen, Sungbin Oh, Isobel Ojalvo, Hideyoshi Ozaki, Simone Pagan Griso, Santosh Parajuli, Myeonghun Park, Benjamin Parpillon, Christoph Paus, Sara Pozzi, Ema Puljak, Benjamin Ramhorst, Amy Roberts, Larry Ruckman, Sebastian Schmitt, Kate Scholberg, Noah Singer, Eluned Anne Smith, Alexandre Sousa, Michael Spannowsky, Kaito Sugizaki, Sioni Summers, Yanwen Sun, Daniel Tapia Takaki, Antonino Tumeo, Caterina Vernieri, Belina von Krosigk, Yash Vora, Linyan Wan, Michael H. L. S. Wang, Amanda Weinstein, Andy White, Simon Williams, Liangyu Wu, Felix Yu

Audio and Speech Processing · Cross-listed from cs.SD

MIDI-Informed Singing Accompaniment Generation in a Compositional Song Pipeline

Fang-Duo Tsai, Yi-An Lai, Fei-Yueh Chen et al. (+4 more)

While end-to-end lyrics-to-song models offer convenience for casual users, professional songwriters require score-to-song systems that allow them to retain authorship over the core melody. However, existing score-to-song methods are limited to short-form snippets and fail to maintain coherence in long-form generation, particularly during vocal-silent sections like intros and bridges. To address this long-form bottleneck, we propose MIDI-informed singing accompaniment generation (MIDI-SAG). Unlike conventional audio-only models, MIDI-SAG utilizes symbolic timing and chord information derived from the vocal MIDI to provide a stable musical roadmap. By incorporating structure planning, which defines temporal boundaries and semantic labels, our framework facilitates consistent generation across both vocal and non-vocal sections. We demonstrate the feasibility of this compositional pipeline by leveraging specialized pre-trained modules, enabling data-efficient training on a single GPU. Our experiments show the potential of this approach for both professional score-to-song and general lyrics-to-song tasks. While an early exploration, MIDI-SAG suggests a promising direction for structured, long-form music synthesis. Audio demos are available, and the code will be open-sourced at https://composerflow.github.io/web_revealed/.
Read the full abstract

All authors Fang-Duo Tsai, Yi-An Lai, Fei-Yueh Chen, Hsueh-Wei Fu, Wei-Jaw Lee, Hao-Chung Cheng, Yi-Hsuan Yang

Systems and Control · Cross-listed from cs.LG

Provably Safe Generative Sampling with Constricting Barrier Functions

Darshan Gadginmath, Ahmed Allibhoy and Fabio Pasqualetti

Flow-based generative models, such as diffusion models and flow matching models, have achieved remarkable success in learning complex data distributions. However, a critical gap remains for their deployment in safety-critical domains: the lack of formal guarantees that generated samples will satisfy hard constraints. We propose a safety filtering framework that acts as an online shield for any pre-trained generative model. Our key insight is to cooperate with the generative process rather than override it. We define a constricting safety tube that is relaxed at the initial noise distribution and progressively tightens to the target safe set at the final data distribution, mirroring the coarse-to-fine structure of the generative process itself. By characterizing this tube via Control Barrier Functions (CBFs), we synthesize a feedback control input through a convex Quadratic Program (QP) at each sampling step. As the tube is loosest when noise is high and intervention is cheapest in terms of control energy, most constraint enforcement occurs when it least disrupts the model's learned structure. We prove that this mechanism guarantees safe sampling in discrete-time. The minimum-norm control synthesized at each step minimizes the per-step contribution to the KL divergence between the safe and original distributions. Across all experiments, we observe 100% constraint satisfaction. Our framework applies to any pre-trained flow-based sampling scheme requiring no retraining or architectural modifications. We validate the approach across constrained image generation, physically-consistent trajectory sampling, and safe robotic manipulation policies, achieving 100% constraint satisfaction while preserving semantic fidelity.
Read the full abstract

Systems and Control · Cross-listed from cs.LG

Benchmarking State Space Models, Transformers, and Recurrent Networks for US Grid Forecasting

Sunki Hong and Jisoo Lee

Selecting the right deep learning model for power grid forecasting is challenging, as performance heavily depends on the data available to the operator. This paper presents a comprehensive benchmark of five modern neural architectures: two state space models (PowerMamba, S-Mamba), two Transformers (iTransformer, PatchTST), and a traditional LSTM. We evaluate these models on hourly electricity demand across six diverse US power grids for forecast windows between 24 and 168 hours. To ensure a fair comparison, we adapt each model with specialized temporal processing and a modular layer that cleanly integrates weather covariates. Our results reveal that there is no single best model for all situations. When forecasting using only historical load, PatchTST and the state space models provide the highest accuracy. However, when explicit weather data is added to the inputs, the rankings reverse: iTransformer improves its accuracy three times more efficiently than PatchTST. By controlling for model size, we confirm that this advantage stems from the architecture's inherent ability to mix information across different variables. Extending our evaluation to solar generation, wind power, and wholesale prices further demonstrates that model rankings depend on the forecast task: PatchTST excels on highly rhythmic signals like solar, while state space models are better suited for the chaotic fluctuations of wind and price. Ultimately, this benchmark provides grid operators with actionable guidelines for selecting the optimal forecasting architecture based on their specific data environments.
Read the full abstract

Signal Processing · Cross-listed from stat.ME

An index of effective number of variables for uncertainty and reliability analysis in model selection problems

Luca Martino, Eduardo Morgado and Roberto San Millán-Castillo

An index of an effective number of variables (ENV) is introduced for model selection in nested models. This is the case, for instance, when we have to decide the order of a polynomial function or the number of bases in a nonlinear regression, choose the number of clusters in a clustering problem, or the number of features in a variable selection application (to name few examples). It is inspired by the idea of the maximum area under the curve (AUC). The interpretation of the ENV index is identical to the effective sample size (ESS) indices concerning a set of samples. The ENV index improves {drawbacks of} the elbow detectors described in the literature and introduces different confidence measures of the proposed solution. These novel measures can be also employed jointly with the use of different information criteria, such as the well-known AIC and BIC, or any other model selection procedures. Comparisons with classical and recent schemes are provided in different experiments involving real datasets. Related Matlab code is given.
Read the full abstract

Image and Video Processing · Cross-listed from cs.CV

Towards Controllable Video Synthesis of Routine and Rare OR Events

Dominik Schneider, Lalithkumar Seenivasan, Sampath Rapuri et al. (+8 more)

Purpose: Curating large-scale datasets of operating room (OR) workflow, encompassing rare, safety-critical, or atypical events, remains operationally and ethically challenging. This data bottleneck complicates the development of ambient intelligence for detecting, understanding, and mitigating rare or safety-critical events in the OR. Methods: This work presents an OR video diffusion framework that enables controlled synthesis of rare and safety-critical events. The framework integrates a geometric abstraction module, a conditioning module, and a fine-tuned diffusion model to first transform OR scenes into abstract geometric representations, then condition the synthesis process, and finally generate realistic OR event videos. Using this framework, we also curate a synthetic dataset to train and validate AI models for detecting near-misses of sterile-field violations. Results: In synthesizing routine OR events, our method outperforms off-the-shelf video diffusion baselines, achieving lower FVD/LPIPS and higher SSIM/PSNR in both in- and out-of-domain datasets. Through qualitative results, we illustrate its ability for controlled video synthesis of counterfactual events. An AI model trained and validated on the generated synthetic data achieved a RECALL of 70.13% in detecting near safety-critical events. Finally, we conduct an ablation study to quantify performance gains from key design choices. Conclusion: Our solution enables controlled synthesis of routine and rare OR events from abstract geometric representations. Beyond demonstrating its capability to generate rare and safety-critical scenarios, we show its potential to support the development of ambient intelligence models.
Read the full abstract

All authors Dominik Schneider, Lalithkumar Seenivasan, Sampath Rapuri, Vishalroshan Anil, Aiza Maksutova, Yiqing Shen, Jan Emily Mangulabnan, Hao Ding, Jose L. Porras, Masaru Ishii, Mathias Unberath

Image and Video Processing

RelA-Diffusion: Relativistic Adversarial Diffusion for Multi-Tracer PET Synthesis from Multi-Sequence MRI

Minhui Yu, Yongheng Sun, David S. Lalush et al. (+3 more)

Multi-tracer positron emission tomography (PET) provides critical insights into diverse neuropathological processes such as tau accumulation, neuroinflammation, and $\beta$-amyloid deposition in the brain, making it indispensable for comprehensive neurological assessment. However, routine acquisition of multi-tracer PET is limited by high costs, radiation exposure, and restricted tracer availability. Recent efforts have explored deep learning approaches for synthesizing PET images from structural MRI. While some methods rely solely on T1-weighted MRI, others incorporate additional sequences such as T2-FLAIR to improve pathological sensitivity. However, existing methods often struggle to capture fine-grained anatomical and pathological details, resulting in artifacts and unrealistic outputs. To this end, we propose RelA-Diffusion, a Relativistic Adversarial Diffusion framework for multi-tracer PET synthesis from multi-sequence MRI. By leveraging both T1-weighted and T2-FLAIR scans as complementary inputs, RelA-Diffusion captures richer structural information to guide image generation. To improve synthesis fidelity, we introduce a gradient-penalized relativistic adversarial loss to the intermediate clean predictions of the diffusion model. This loss compares real and generated images in a relative manner, encouraging the synthesis of more realistic local structures. Both the relativistic formulation and the gradient penalty contribute to stabilizing the training, while adversarial feedback at each diffusion timestep enables consistent refinement throughout the generation process. Extensive experiments on two datasets demonstrate that RelA-Diffusion outperforms existing methods in both visual fidelity and quantitative metrics, highlighting its potential for accurate synthesis of multi-tracer PET.
Read the full abstract

All authors Minhui Yu, Yongheng Sun, David S. Lalush, Jason P Mihalik, Pew-Thian Yap, Mingxia Liu

Image and Video Processing

On Optimizing Image Codecs for VMAF NEG: Analysis, Issues, and a Robust Loss Proposal

Florian Fingscheidt, Alexander Karabutov, Panqi Jia et al. (+2 more)

The VMAF (video multi-method assessment fusion) metric for image and video coding recently gained more and more popularity as it is supposed to have a high correlation with human perception. This makes training and particularly fine-tuning of machine-learned codecs on this metric interesting. However, VMAF is shown to be attackable in a way that, e.g., unsharpening an image can lead to a gain in VMAF quality while decreasing the quality in human perception. A particular version of VMAF called VMAF NEG has been designed to be more robust against such attacks and therefore it should be more useful for fine-tuning of codecs. In this paper, our contributions are threefold. First, we identify and analyze the still existing vulnerability of VMAF NEG towards attacks, particulary towards the attack that consists in employing VMAF NEG for image codec fine-tuning. Second, to benefit from VMAF NEG's high correlation with human perception, we propose a robust loss including VMAF NEG for fine-tuning either the encoder or the decoder. Third, we support our quantitative objective results by providing perceptive impressions of some image examples.
Read the full abstract

All authors Florian Fingscheidt, Alexander Karabutov, Panqi Jia, Elena Alshina, Jörn Ostermann

Systems and Control

Autonomous Satellite Rendezvous via Hybrid Feedback Optimization

Oscar Jed R. Chuy, Matthew T. Hale, Vignesh Sivaramakrishnan et al. (+2 more)

As satellites have proliferated, interest has increased in autonomous rendezvous, proximity operations, and docking (ARPOD). A fundamental challenge in these tasks is the uncertainties when operating in space, e.g., in measurements of satellites' states, which can make future states difficult to predict. Another challenge is that satellites' onboard processors are typically much slower than their terrestrial counterparts. Therefore, to address these challenges we propose to solve an ARPOD problem with feedback optimization, which computes inputs to a system by measuring its outputs, feeding them into an optimization algorithm in the loop, and computing some number of iterations towards an optimal input. We focus on satellite rendezvous, and satellites' dynamics are modeled using the continuous-time Clohessy-Wiltshire equations, which are marginally stable. We develop an asymptotically stabilizing controller for them, and we use discrete-time gradient descent in the loop to compute inputs to them. Then, we analyze the hybrid feedback optimization system formed by the stabilized Clohessy-Wiltshire equations with gradient descent in the loop. We show that this model is well-posed and that maximal solutions are both complete and non-Zeno. Then, we show that solutions converge exponentially fast to a ball around a rendezvous point, and we bound the radius of that ball in terms of system parameters. Simulations show that this approach provides up to a 98.4% reduction in the magnitude of disturbances across a range of simulations, which illustrates the viability of hybrid feedback optimization for autonomous satellite rendezvous.
Read the full abstract

All authors Oscar Jed R. Chuy, Matthew T. Hale, Vignesh Sivaramakrishnan, Sean Phillips, Ricardo G. Sanfelice

Systems and Control · Cross-listed from cs.AR

Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler

Xinxin Wang, Lixian Yan, Shuhan Liu et al. (+10 more)

As memory increasingly dominates system cost and energy, heterogeneous on-chip memory systems that combine technologies with complementary characteristics are becoming essential. Gain Cell RAM (GCRAM) offers higher density, lower power, and tunable retention, expanding the design space beyond conventional SRAM. To this end, we create an OpenGCRAM compiler supporting both SRAM and GCRAM. It generates macro-level designs and layouts for commercial CMOS processes and characterizes area, delay, and power across user-defined configurations. The tool enables systematic identification of optimal heterogeneous memory configurations for AI tasks under specified performance metrics.
Read the full abstract

All authors Xinxin Wang, Lixian Yan, Shuhan Liu, Luke Upton, Zhuoqi Cai, Yiming Tan, Shengman Li, Koustav Jana, Peijing Li, Jesse Cirimelli-Low, Thierry Tambe, Matthew Guthaus, H.-S. Philip Wong

Image and Video Processing

A Light Fixture Color Temperature and Color Rendering Index Measuring Device

Gianluca Hiss Garbim, Luis Carlos Mathias, André Massami Assakawa et al. (+1 more)

The correlated color temperature (CCT) and color rendering index (CRI) of artificial light sources are important because they have implications for human biology and professional applications. Although CCT information is generally available for commercial lamps, CRI is commonly not reported. In addition, devices measuring these parameters are difficult to access as they require a spectrophotometer, a commonly expensive device. In this context, the present work designs and builds a meter in detail, from the structural part of the equipment, interface with sensors, and the calculation to the compensation algorithms implementation, aiming to build the dedicated functionalities of a spectrophotometer, which is designed without the use of optical lenses. In addition to simplifying the device, this approach allows measurements free from dispersions caused by chromatic aberrations typical of optical lenses. The prototype obtained proved to be effective, capturing the spectral power distributions of various light sources and calculating their CCT and CRI.
Read the full abstract

All authors Gianluca Hiss Garbim, Luis Carlos Mathias, André Massami Assakawa, Taufik Abrão

Systems and Control

KAN-Koopman Based Rapid Detection Of Battery Thermal Anomalies With Diagnostics Guarantees

Sanchita Ghosh and Tanushree Roy

Early diagnosis of battery thermal anomalies is crucial to ensure safe and reliable battery operation by preventing catastrophic thermal failures. Battery diagnostics primarily rely on battery surface temperature measurements and/or estimation of core temperatures. However, aging-induced changes in the battery model and limited training data remain major challenges for model-based and machine-learning based battery state estimation and diagnostics. To address these issues, we propose a Kolomogorov-Arnold network (KAN) in conjunction with a Koopman-based detection algorithm that leverages the unique advantages of both methods. Firstly, the lightweight KAN provides a model-free estimation of the core temperature to ensure rapid detection of battery thermal anomalies. Secondly, the Koopman operator is learned in real time using the estimated core temperature from KAN and the measured surface temperature of the battery to provide the core and surface temperature prediction for diagnostic residual generation. This online learning approach overcomes the challenges of model changes. Furthermore, we derive analytical conditions to obtain diagnostic guarantees on our KAN-Koopman detection scheme. Our simulation results illustrate a significant reduction in detection time with the proposed algorithm compared to the baseline Koopman-only algorithm.
Read the full abstract

Image and Video Processing

Vision-Inspired Image Quality Assessment for Radar-Based Human Activity Representations

Huy Trinh, Davis Liu, Munia Humaira et al. (+2 more)

Radar-based human activity recognition has gained attention as a privacy-preserving alternative to vision and wearable sensors, especially in sensitive environments like long-term care facilities. Micro-Doppler spectrograms derived from FMCW radar signals are central to recognizing dynamic activities, but their effectiveness is limited by noise and clutter. In this work, we use a benchmark radar dataset to reimplement and assess three recent denoising and preprocessing techniques: adaptive preprocessing, adaptive thresholding, and entropy-based denoising. To illustrate the shortcomings of conventional metrics in low-SNR regimes, we evaluate performance using both perceptual image quality measures and standard error-based metrics. We additionally propose a novel framework for static activity recognition using range-angle feature maps to expand HAR beyond dynamic activities. We present two important contributions: a temporal tracking algorithm to enforce consistency and a no-reference quality scoring algorithm to assess RA-map fidelity. According to experimental findings, our suggested techniques enhance classification performance and interpretability for both dynamic and static activities, opening the door for more reliable radar-based HAR systems.
Read the full abstract

All authors Huy Trinh, Davis Liu, Munia Humaira, Peter Lee, Zhou Wang

Signal Processing

Attention-Based SINR Estimation in User-Centric Non-Terrestrial Networks

Bruno De Filippo, Alessandro Guidotti and Alessandro Vanelli-Coralli

The signal-to-interference-plus-noise ratio (SINR) is central to performance optimization in user-centric beamforming for satellite-based non-terrestrial networks (NTNs). Its assessment either requires the transmission of dedicated pilots or relies on computing the beamforming matrix through minimum mean squared error (MMSE)-based formulations beforehand, a process that introduces significant computational overhead. In this paper, we propose a low-complexity SINR estimation framework that leverages multi-head self-attention (MHSA) to extract inter-user interference features directly from either channel state information or user location reports. The proposed dual MHSA (DMHSA) models evaluate the SINR of a scheduled user group without requiring explicit MMSE calculations. The architecture achieves a computational complexity reduction by a factor of three in the CSI-based setting and by two orders of magnitude in the location-based configuration, the latter benefiting from the lower dimensionality of user reports. We show that both DMHSA models maintain high estimation accuracy, with the root mean squared error typically below 1 dB with priority-queuing-based scheduled users. These results enable the integration of DMHSA-based estimators into scheduling procedures, allowing the evaluation of multiple candidate user groups and the selection of those offering the highest average SINR and capacity.
Read the full abstract

Systems and Control

A Survey of Recent Developments in SYCL Compiler Implementations

Huy Trinh

This survey discusses recent advancements in SYCL compiler implementations, one of the crucial aspects of compiler construction for heterogeneous computing systems. We explore the transition from traditional compiler construction, from Single-Source Multiple Compiler Passes (SMCP) to a more advanced approach to Single-Source Single Compiler Pass (SSCP). The survey analyzes multiple papers that researched the different developments of SYCL implementation based on SSCP and their approach to enhancing performance and addressing separate challenges.
Read the full abstract