Tootfinder

Heart Disease Prediction: A Comparative Study of Optimisers Performance in Deep Neural Networks
Chisom Chibuike, Adeyinka Ogunsanya
https://arxiv.org/abs/2509.08499 https://

Heart Disease Prediction: A Comparative Study of Optimisers Performance in Deep Neural Networks
Optimization has been an important factor and topic of interest in training deep learning models, yet less attention has been given to how we select the optimizers we use to train these models. Hence, there is a need to dive deeper into how we select the optimizers we use for training and the metrics that determine this selection. In this work, we compare the performance of 10 different optimizers in training a simple Multi-layer Perceptron model using a heart disease dataset from Kaggle. We se…

Hyperspectral data augmentation with transformer-based diffusion models
Mattia Ferrari, Lorenzo Bruzzone
https://arxiv.org/abs/2510.08363 https://arxiv.org…

Hyperspectral data augmentation with transformer-based diffusion models
The introduction of new generation hyperspectral satellite sensors, combined with advancements in deep learning methodologies, has significantly enhanced the ability to discriminate detailed land-cover classes at medium-large scales. However, a significant challenge in deep learning methods is the risk of overfitting when training networks with small labeled datasets. In this work, we propose a data augmentation technique that leverages a guided diffusion model. To effectively train the model w…

Application of LLMs to Multi-Robot Path Planning and Task Allocation
Ashish Kumar
https://arxiv.org/abs/2507.07302 https://arxiv.org/…

Application of LLMs to Multi-Robot Path Planning and Task Allocation
Efficient exploration is a well known problem in deep reinforcement learning and this problem is exacerbated in multi-agent reinforcement learning due the intrinsic complexities of such algorithms. There are several approaches to efficiently explore an environment to learn to solve tasks by multi-agent operating in that environment, of which, the idea of expert exploration is investigated in this work. More specifically, this work investigates the application of large-language models as expert …

Enhanced SegNet with Integrated Grad-CAM for Interpretable Retinal Layer Segmentation in OCT Images
S M Asiful Islam Saky, Ugyen Tshering
https://arxiv.org/abs/2509.07795 https:…

Enhanced SegNet with Integrated Grad-CAM for Interpretable Retinal Layer Segmentation in OCT Images
Optical Coherence Tomography (OCT) is essential for diagnosing conditions such as glaucoma, diabetic retinopathy, and age-related macular degeneration. Accurate retinal layer segmentation enables quantitative biomarkers critical for clinical decision-making, but manual segmentation is time-consuming and variable, while conventional deep learning models often lack interpretability. This work proposes an improved SegNet-based deep learning framework for automated and interpretable retinal layer s…

Decoding the dark proteome: Deep learning-enabled discovery of druggable enzymes in Wuchereria bancrofti
Shawnak Shivakumar, Jefferson Hernandez
https://arxiv.org/abs/2510.07337

Decoding the dark proteome: Deep learning-enabled discovery of druggable enzymes in Wuchereria bancrofti
Wuchereria bancrofti, the parasitic roundworm responsible for lymphatic filariasis, permanently disables over 36 million people and places 657 million at risk across 39 countries. A major bottleneck for drug discovery is the lack of functional annotation for more than 90 percent of the W. bancrofti dark proteome, leaving many potential targets unidentified. In this work, we present a novel computational pipeline that converts W. bancrofti's unannotated amino acid sequence data into precise four…

Automated Trading System for Straddle-Option Based on Deep Q-Learning
Yiran Wan, Xinyu Ying, Shengzhen Xu
https://arxiv.org/abs/2509.07987 https://arxiv.or…

Automated Trading System for Straddle-Option Based on Deep Q-Learning
Straddle Option is a financial trading tool that explores volatility premiums in high-volatility markets without predicting price direction. Although deep reinforcement learning has emerged as a powerful approach to trading automation in financial markets, existing work mostly focused on predicting price trends and making trading decisions by combining multi-dimensional datasets like blogs and videos, which led to high computational costs and unstable performance in high-volatility markets. To …

Reduction Techniques for Survival Analysis
Johannes Piller, L\'ea Orsini, Simon Wiegrebe, John Zobolas, Lukas Burk, Sophie Hanna Langbein, Philip Studener, Markus Goeswein, Andreas Bender
https://arxiv.org/abs/2508.05715

Reduction Techniques for Survival Analysis
In this work, we discuss what we refer to as reduction techniques for survival analysis, that is, techniques that "reduce" a survival task to a more common regression or classification task, without ignoring the specifics of survival data. Such techniques particularly facilitate machine learning-based survival analysis, as they allow for applying standard tools from machine and deep learning to many survival tasks without requiring custom learners. We provide an overview of different reduction …

Day 16: Mayra Cuevas & Marie Marquardt
Okay so this is cheating, but they're co-authors of multiple books together, and there's no way for me to separate their contributions... I've already got too many authors I'd like to list, so why not?
I read their book "Does My Body Offend You?" and absolutely loved it; it's a celebration of teen activism while also being a deep exploration of feminist issues through practical situations that bring out the complicated side of things, which the authors refuse to reduce back to a simple formulaic answer. It has a supporting cast of appropriately-complex male characters that help in exploring the nuances of issues like the line between female empowerment & male gratification, and it brings race and macho culture into the conversion as well.
CW for sexual harassment & deep discussion of the resultant trauma.
I'll cheat again here to sneak in mention of two male authors whose work resonates with theirs: Mark Oshiro's "Anger is a Gift" has a more pessimistic/complex take on teen activism along with a gay romance (CW for racist cop murder), while Jeremy Whitley's graphic novel "Navigating With You" deals with queer romance & disability, while having a main character pairing that echoes those from "Does My Body Offend You?" in a lot of ways. Another connection (to non-men authors this time) is with "Go With the Flow" by Lily Williams and Karen Schneemann. Their graphic novel about teen activism and periods is a bit more didactic and has a much lighter tone, but it does necessarily have some overlapping themes.
To bring it back to Cuevas & Marquhardt, their writing is great and their ability to discuss such complex topics with such nuance, all wrapped up in a story that feels completely natural, is amazing to me, and makes their book feel like one of the most valuable to recommend to others.
In writing this I've realized a grave oversight in the list so far that I'll have to correct tomorrow, but I'm quickly running out of days. The didn't-quite-make-it list is going to be full of more excellent authors, and I'm honestly starting to wonder whether it might actually be harder to name 20 male authors I respect now that I've found the sense to be mostly somewhere between disgusted and disappointed with so many of the male authors I enjoyed as a teen.
#20AuthorsNoMen (cheating a bit)

A Scalable FPGA Architecture With Adaptive Memory Utilization for GEMM-Based Operations
Anastasios Petropoulos, Theodore Antonakopoulos
https://arxiv.org/abs/2510.08137 https://…

A Scalable FPGA Architecture With Adaptive Memory Utilization for GEMM-Based Operations
Deep neural network (DNN) inference relies increasingly on specialized hardware for high computational efficiency. This work introduces a field-programmable gate array (FPGA)-based dynamically configurable accelerator featuring systolic arrays, high-bandwidth memory, and UltraRAMs. We present two processing unit (PU) configurations with different computing capabilities using the same interfaces and peripheral blocks. By instantiating multiple PUs and employing a heuristic weight transfer schedu…

Universal Graph Learning for Power System Reconfigurations: Transfer Across Topology Variations
Tong Wu, Anna Scaglione, Sandy Miguel, Daniel Arnold
https://arxiv.org/abs/2509.08672

Universal Graph Learning for Power System Reconfigurations: Transfer Across Topology Variations
This work addresses a fundamental challenge in applying deep learning to power systems: developing neural network models that transfer across significant system changes, including networks with entirely different topologies and dimensionalities, without requiring training data from unseen reconfigurations. Despite extensive research, most ML-based approaches remain system-specific, limiting real-world deployment. This limitation stems from a dual barrier. First, topology changes shift feature d…

Neural Proxies for Sound Synthesizers: Learning Perceptually Informed Preset Representations
Paolo Combes, Stefan Weinzierl, Klaus Obermayer
https://arxiv.org/abs/2509.07635 htt…

Neural Proxies for Sound Synthesizers: Learning Perceptually Informed Preset Representations
Deep learning appears as an appealing solution for Automatic Synthesizer Programming (ASP), which aims to assist musicians and sound designers in programming sound synthesizers. However, integrating software synthesizers into training pipelines is challenging due to their potential non-differentiability. This work tackles this challenge by introducing a method to approximate arbitrary synthesizers. Specifically, we train a neural network to map synthesizer presets onto an audio embedding space …

Modular PE-Structured Learning for Cross-Task Wireless Communications
Yuxuan Duan, Chenyang Yang
https://arxiv.org/abs/2509.08614 https://arxiv.org/pdf/250…

Modular PE-Structured Learning for Cross-Task Wireless Communications
Recent trends in learning wireless policies attempt to develop deep neural networks (DNNs) for handling multiple tasks with a single model. Existing approaches often rely on large models, which are hard to pre-train and fine-tune at the wireless edge. In this work, we challenge this paradigm by leveraging the structured knowledge of wireless problems -- specifically, permutation equivariant (PE) properties. We design three types of PE-aware modules, two of which are Transformer-style sub-layers…

Accelerating Inference for Multilayer Neural Networks with Quantum Computers
Arthur G. Rattew, Po-Wei Huang, Naixu Guo, Lirand\"e Pira, Patrick Rebentrost
https://arxiv.org/abs/2510.07195

Accelerating Inference for Multilayer Neural Networks with Quantum Computers
Fault-tolerant Quantum Processing Units (QPUs) promise to deliver exponential speed-ups in select computational tasks, yet their integration into modern deep learning pipelines remains unclear. In this work, we take a step towards bridging this gap by presenting the first fully-coherent quantum implementation of a multilayer neural network with non-linear activation functions. Our constructions mirror widely used deep learning architectures based on ResNet, and consist of residual blocks with m…

Dark Energy Survey Year 6 Results: Redshift Calibration of the MagLim Lens Sample
G. Giannini, A. Alarcon, W. d'Assignies, G. M. Bernstein, M. A. Troxel, C. Chang, B. Yin, A. Amon, J. Myles, N. Weaverdyck, A. Porredon, D. Anbajagane, S. Avila, K. Bechtol, M. R. Becker, J. Blazek, M. Crocce, D. Gruen, M. Rodriguez-Monroy, C. S\'anchez, D. Sanchez Cid, I. Sevilla-Noarbe, M. Aguena, S. Allam, O. Alves, F. Andrade-Oliveira, D. Bacon, S. Bocquet, D. Brooks, R. Camilleri, A. Carner…

Dark Energy Survey Year 6 Results: Redshift Calibration of the MagLim++ Lens Sample
In this work, we derive and calibrate the redshift distribution of the MagLim++ lens galaxy sample used in the Dark Energy Survey Year 6 (DES Y6) 3x2pt cosmology analysis. The 3x2pt analysis combines galaxy clustering from the lens galaxy sample and weak gravitational lensing. The redshift distributions are inferred using the SOMPZ method - a Self-Organizing Map framework that combines deep-field multi-band photometry, wide-field data, and a synthetic source injection (Balrog) catalog. Key impr…

Learning Polynomial Activation Functions for Deep Neural Networks
Linghao Zhang, Jiawang Nie, Tingting Tang
https://arxiv.org/abs/2510.03682 https://arxiv.…

Learning Polynomial Activation Functions for Deep Neural Networks
Activation functions are crucial for deep neural networks. This novel work frames the problem of training neural network with learnable polynomial activation functions as a polynomial optimization problem, which is solvable by the Moment-SOS hierarchy. This work represents a fundamental departure from the conventional paradigm of training deep neural networks, which relies on local optimization methods like backpropagation and gradient descent. Numerical experiments are presented to demonstrate…

ACE and Diverse Generalization via Selective Disagreement
Oliver Daniels, Stuart Armstrong, Alexandre Maranh\~ao, Mahirah Fairuz Rahman, Benjamin M. Marlin, Rebecca Gorman
https://arxiv.org/abs/2509.07955

ACE and Diverse Generalization via Selective Disagreement
Deep neural networks are notoriously sensitive to spurious correlations - where a model learns a shortcut that fails out-of-distribution. Existing work on spurious correlations has often focused on incomplete correlations,leveraging access to labeled instances that break the correlation. But in cases where the spurious correlations are complete, the correct generalization is fundamentally \textit{underspecified}. To resolve this underspecification, we propose learning a set of concepts that are…

The Small Magellanic Cloud through the lens of the James Webb Space Telescope : binaries and mass function within the galaxy outskirts
M. V. Legnardi, F. Muratore, A. P. Milone, G. Cordoni, T. Ziliotto, E. Dondoglio, A. F. Marino, A. Mastrobuono-Battisti, E. Bortolan, E. P. Lagioia, M. Tailo
https://arxiv.org/abs/2509.08687

The Small Magellanic Cloud through the lens of the James Webb Space Telescope : binaries and mass function within the galaxy outskirts
The stellar initial mass function (IMF) and the fraction of binary systems are fundamental ingredients that govern the formation and evolution of galaxies. Whether the IMF is universal or varies with environment remains one of the central open questions in astrophysics. Dwarf galaxies such as the Small Magellanic Cloud (SMC), with their low metallicity and diffuse star-forming regions, offer critical laboratories to address this issue. In this work, we exploit ultra-deep photometry from the Jam…

Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
Xin Lai, Junyi Li, Wei Li, Tao Liu, Tianjian Li, Hengshuang Zhao
https://arxiv.org/abs/2509.07969 …

Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
Recent advances in large multimodal models have leveraged image-based tools with reinforcement learning to tackle visual problems. However, existing open-source approaches often exhibit monotonous reasoning patterns and allow only a limited number of interaction turns, making them inadequate for difficult tasks that require trial-and-error exploration. In this work, we address this limitation by scaling up tool-based interactions and introduce Mini-o3, a system that executes deep, multi-turn re…

Deep Inverse Rosenblatt Transport for Structural Reliability Analysis
Aryan Tyagi, Jan N. Fuhg
https://arxiv.org/abs/2509.05061 https://arxiv.org/pdf/2509.…

Deep Inverse Rosenblatt Transport for Structural Reliability Analysis
Accurately estimating the probability of failure in engineering systems under uncertainty is a fundamental challenge, particularly in high-dimensional settings and for rare events. Conventional reliability analysis methods often become computationally intractable or exhibit high estimator variance when applied to problems with hundreds of uncertain parameters or highly concentrated failure regions. In this work, we investigate the use of the recently proposed Deep Inverse Rosenblatt Transport (…

Vertex reconstruction in the TAO experiment
Hangyu Shi, Jun Wang, Guofu Cao, Wei Wang, Yuehuan Wei
https://arxiv.org/abs/2508.06293 https://arxiv.org/pdf/2…

Vertex reconstruction in the TAO experiment
The Taishan Antineutrino Observatory (TAO) is a tonne-scale gadolinium-doped liquid scintillator satellite experiment of the Jiangmen Underground Neutrino Observatory (JUNO). It is designed to measure the reactor antineutrino energy spectrum with unprecedented energy resolution, better than 2% at 1 MeV. To fully achieve its designed performance, precise vertex reconstruction is crucial. This work reports two distinct vertex reconstruction methods, the charge center algorithm (CCA) and the deep …

A Deep Q-Network based power control mechanism to Minimize RLF driven Handover Failure in 5G Network
Kotha Kartheek, Shankar K. Ghosh, Megha Iyengar, Vinod Sharma, Souvik Deb
https://arxiv.org/abs/2510.05762

A Deep Q-Network based power control mechanism to Minimize RLF driven Handover Failure in 5G Network
The impact of Radio link failure (RLF) has been largely ignored in designing handover algorithms, although RLF is a major contributor towards causing handover failure (HF). RLF can cause HF if it is detected during an ongoing handover. The objective of this work is to propose an efficient power control mechanism based on Deep Q-Network (DQN), considering handover parameters (i.e., time-to-preparation, time-to-execute, preparation offset, execution offset) and radio link monitoring parameters (T…

Langtauferer
Today, a year ago, I ventured on a dream trip I'd been researching for a long time, and which ended up being a semi-religious experience, being immersed in (and somewhat overwhelmed by) an actively changing environment, the upheaval and plethora of geological features, structures, unreal colors, layers, textures and the "wounds" exposed by the melting and disappearing glaciers... Countless waterfalls, stunning erosion features, later traversing the glacier ga…

QDeepGR4J: Quantile-based ensemble of deep learning and GR4J hybrid rainfall-runoff models for extreme flow prediction with uncertainty quantification
Arpit Kapoor, Rohitash Chandra
https://arxiv.org/abs/2510.05453

QDeepGR4J: Quantile-based ensemble of deep learning and GR4J hybrid rainfall-runoff models for extreme flow prediction with uncertainty quantification
Conceptual rainfall-runoff models aid hydrologists and climate scientists in modelling streamflow to inform water management practices. Recent advances in deep learning have unravelled the potential for combining hydrological models with deep learning models for better interpretability and improved predictive performance. In our previous work, we introduced DeepGR4J, which enhanced the GR4J conceptual rainfall-runoff model using a deep learning model to serve as a surrogate for the routing comp…

How long can you sleep? Idle Time System Inefficiencies and Opportunities
Georgia Antoniou (University of Cyprus), Haris Volos (University of Cyprus), Jawad Haj Yahya (Rivos Inc), Yiannakis Sazeides (University of Cyprus)
https://arxiv.org/abs/2510.07449

How long can you sleep? Idle Time System Inefficiencies and Opportunities
This work introduces a model-based framework that reveals the idle opportunity of modern servers running latency-critical applications. Specifically, three queuing models, M/M/1, cxM/M/1, and M/M/c, are used to estimate the theoretical idle time distribution at the CPU core and system (package) level. A comparison of the actual idleness of a real server and that from the theoretical models reveals significant missed opportunities to enter deep idle states. This inefficiency is attributed to the…

Audio-Visual Speech Separation via Bottleneck Iterative Network
Sidong Zhang, Shiv Shankar, Trang Nguyen, Andrea Fanelli, Madalina Fiterau
https://arxiv.org/abs/2507.07270

Audio-Visual Speech Separation via Bottleneck Iterative Network
Integration of information from non-auditory cues can significantly improve the performance of speech-separation models. Often such models use deep modality-specific networks to obtain unimodal features, and risk being too costly or lightweight but lacking capacity. In this work, we present an iterative representation refinement approach called Bottleneck Iterative Network (BIN), a technique that repeatedly progresses through a lightweight fusion block, while bottlenecking fusion representation…

Sparse Seemingly Unrelated Regression (SSUR) Copula Mixed Models for Multivariate Loss Reserving
Pengfei Cai, Anas Abdallah, Pratheepa Jeganathan
https://arxiv.org/abs/2509.05426

Sparse Seemingly Unrelated Regression (SSUR) Copula Mixed Models for Multivariate Loss Reserving
Insurance companies often operate across multiple interrelated lines of business (LOBs), and accounting for dependencies between them is essential for accurate reserve estimation and risk capital determination. In our previous work on the Extended Deep Triangle (EDT), we demonstrated that a more flexible model that uses multiple companies' data reduces reserve prediction error and increases diversification benefits. However, the EDT's limitation lies in its limited interpretability of the depen…

CSI Compression Beyond Latents: End-to-End Hybrid Attention-CNN Networks with Entropy Regularization
Maryam Ansarifard, Mostafa Rahmani, Mohit K. Sharma, Kishor C. Joshi, George Exarchakos, Alister Burr
https://arxiv.org/abs/2509.08776

CSI Compression Beyond Latents: End-to-End Hybrid Attention-CNN Networks with Entropy Regularization
Massive MIMO systems rely on accurate Channel State Information (CSI) feedback to enable high-gain beam- forming. However, the feedback overhead scales linearly with the number of antennas, presenting a major bottleneck. While recent deep learning methods have improved CSI compres- sion, most overlook the impact of quantization and entropy coding, limiting their practical deployability. In this work, we propose an end-to-end CSI compression framework that inte- grates a Spatial Correlation-Guid…

IP-Basis PINNs: Efficient Multi-Query Inverse Parameter Estimation
Shalev Manor, Mohammad Kohandel
https://arxiv.org/abs/2509.07245 https://arxiv.org/pdf/2…

IP-Basis PINNs: Efficient Multi-Query Inverse Parameter Estimation
Solving inverse problems with Physics-Informed Neural Networks (PINNs) is computationally expensive for multi-query scenarios, as each new set of observed data requires a new, expensive training procedure. We present Inverse-Parameter Basis PINNs (IP-Basis PINNs), a meta-learning framework that extends the foundational work of Desai et al. (2022) to enable rapid and efficient inference for inverse problems. Our method employs an offline-online decomposition: a deep network is first trained offl…

Toward Accessible Dermatology: Skin Lesion Classification Using Deep Learning Models on Mobile-Acquired Images
Asif Newaz, Masum Mushfiq Ishti, A Z M Ashraful Azam, Asif Ur Rahman Adib
https://arxiv.org/abs/2509.04800

Toward Accessible Dermatology: Skin Lesion Classification Using Deep Learning Models on Mobile-Acquired Images
Skin diseases are among the most prevalent health concerns worldwide, yet conventional diagnostic methods are often costly, complex, and unavailable in low-resource settings. Automated classification using deep learning has emerged as a promising alternative, but existing studies are mostly limited to dermoscopic datasets and a narrow range of disease classes. In this work, we curate a large dataset of over 50 skin disease categories captured with mobile devices, making it more representative o…

Audubon welcomes visionary environmentalist Mark Collins to its Board of Directors, bringing financial expertise and a commitment to conservation to the organization's efforts to protect habitats and address climate change. #climatechange #climatesolutions

Audubon Adds Visionary Environmentalist to Board
Mark Collins brings deep experience in environmental causes to guide Audubon’s work to address the dual crises of biodiversity loss and climate change.

Gaussian Equivalence for Self-Attention: Asymptotic Spectral Analysis of Attention Matrix
Tomohiro Hayase, Beno\^it Collins, Ryo Karakida
https://arxiv.org/abs/2510.06685 https:…

Gaussian Equivalence for Self-Attention: Asymptotic Spectral Analysis of Attention Matrix
Self-attention layers have become fundamental building blocks of modern deep neural networks, yet their theoretical understanding remains limited, particularly from the perspective of random matrix theory. In this work, we provide a rigorous analysis of the singular value spectrum of the attention matrix and establish the first Gaussian equivalence result for attention. In a natural regime where the inverse temperature remains of constant order, we show that the singular value distribution of t…

Deep Learning-Driven Prediction of Microstructure Evolution via Latent Space Interpolation
Sachin Gaikwad, Thejas Kasilingam, Owais Ahmad, Rajdip Mukherjee, Somnath Bhowmick
https://arxiv.org/abs/2508.01822

Deep Learning-Driven Prediction of Microstructure Evolution via Latent Space Interpolation
Phase-field models accurately simulate microstructure evolution, but their dependence on solving complex differential equations makes them computationally expensive. This work achieves a significant acceleration via a novel deep learning-based framework, utilizing a Conditional Variational Autoencoder (CVAE) coupled with Cubic Spline Interpolation and Spherical Linear Interpolation (SLERP). We demonstrate the method for binary spinodal decomposition by predicting microstructure evolution for in…

Painting the market: generative diffusion models for financial limit order book simulation and forecasting
Alfred Backhouse, Kang Li, Jakob Foerster, Anisoara Calinescu, Stefan Zohren
https://arxiv.org/abs/2509.05107

Painting the market: generative diffusion models for financial limit order book simulation and forecasting
Simulating limit order books (LOBs) has important applications across forecasting and backtesting for financial market data. However, deep generative models struggle in this context due to the high noise and complexity of the data. Previous work uses autoregressive models, although these experience error accumulation over longer-time sequences. We introduce a novel approach, converting LOB data into a structured image format, and applying diffusion models with inpainting to generate future LOB …

Extreme anisotropies in deep layers of an exploding star: overabundance of Cr in the northeastern jet of Cassiopeia A
Vincenzo Sapienza, Marco Miceli, Masaomi Ono, Shigehiro Nagataki, Takashi Yoshida, Emanuele Greco, Salvatore Orlando, Fabrizio Bocchino
https://arxiv.org/abs/2508.00376

Extreme anisotropies in deep layers of an exploding star: overabundance of Cr in the northeastern jet of Cassiopeia A
Core-collapse supernovae drive nucleosynthesis under extreme thermodynamic conditions, and complex mechanisms are at work prompting the transport of heavy elements from deep stellar interiors into outer layers. We present spatially resolved X-ray spectroscopy of Cassiopeia A's (Cas A) northeastern (NE) jet using the archival 1 Ms Chandra/ACIS observations, and focusing on three fingers of the jet. We report the highest Cr/Fe mass ratio (Cr/Fe $\sim0.14$) ever observed in Cas A, localized in a c…

FASL-Seg: Anatomy and Tool Segmentation of Surgical Scenes
Muraam Abdel-Ghani, Mahmoud Ali, Mohamed Ali, Fatmaelzahraa Ahmed, Mohamed Arsalan, Abdulaziz Al-Ali, Shidin Balakrishnan
https://arxiv.org/abs/2509.06159

FASL-Seg: Anatomy and Tool Segmentation of Surgical Scenes
The growing popularity of robotic minimally invasive surgeries has made deep learning-based surgical training a key area of research. A thorough understanding of the surgical scene components is crucial, which semantic segmentation models can help achieve. However, most existing work focuses on surgical tools and overlooks anatomical objects. Additionally, current state-of-the-art (SOTA) models struggle to balance capturing high-level contextual features and low-level edge features. We propose …

Identifying Microlensing by Compact Dark Matter through Diffraction Patterns in Gravitational Waves with Machine Learning
Ao Liu, Tonghua Liu, Dejiang Li, Cuihong Wen, Jieci Wang, Kai Liao, Jiaxing Cui, Huan Zhou
https://arxiv.org/abs/2509.04538

Identifying Microlensing by Compact Dark Matter through Diffraction Patterns in Gravitational Waves with Machine Learning
Gravitational wave lensing, particularly microlensing by compact dark matter (DM), offers a unique avenue to probe the nature of dark matter. However, conventional detection methods are often computationally expensive, inefficient, and sensitive to waveform systematics. In this work, we introduce the Wavelet Convolution Detector (WCD), a deep learning framework specifically designed to identify wave-optics diffraction patterns imprinted in gravitationally lensed signals. The WCD integrates mult…

One-weight codes in the sum-rank metric
Usman Mushrraf, Ferdinando Zullo
https://arxiv.org/abs/2508.04262 https://arxiv.org/pdf/2508.04262

One-weight codes in the sum-rank metric
One-weight codes, in which all nonzero codewords share the same weight, form a highly structured class of linear codes with deep connections to finite geometry. While their classification is well understood in the Hamming and rank metrics - being equivalent to (direct sums of) simplex codes - the sum-rank metric presents a far more intricate landscape. In this work, we explore the geometry of one-weight sum-rank metric codes, focusing on three distinct classes. First, we introduce and classify …

Deep Fuzzy Optimization for Batch-Size and Nearest Neighbors in Optimal Robot Motion Planning
Liding Zhang, Qiyang Zong, Yu Zhang, Zhenshan Bing, Alois Knoll
https://arxiv.org/abs/2508.20884

Deep Fuzzy Optimization for Batch-Size and Nearest Neighbors in Optimal Robot Motion Planning
Efficient motion planning algorithms are essential in robotics. Optimizing essential parameters, such as batch size and nearest neighbor selection in sampling-based methods, can enhance performance in the planning process. However, existing approaches often lack environmental adaptability. Inspired by the method of the deep fuzzy neural networks, this work introduces Learning-based Informed Trees (LIT*), a sampling-based deep fuzzy learning-based planner that dynamically adjusts batch size and …

On the Evaluation of Large Language Models in Multilingual Vulnerability Repair
Dong wang, Junji Yu, Honglin Shu, Michael Fu, Chakkrit Tantithamthavorn, Yasutaka Kamei, Junjie Chen
https://arxiv.org/abs/2508.03470

On the Evaluation of Large Language Models in Multilingual Vulnerability Repair
Various Deep Learning-based approaches with pre-trained language models have been proposed for automatically repairing software vulnerabilities. However, these approaches are limited to a specific programming language (C/C++). Recent advances in large language models (LLMs) offer language-agnostic capabilities and strong semantic understanding, exhibiting potential to overcome multilingual vulnerability limitations. Although some work has begun to explore LLMs' repair performance, their effecti…

Tenyidie Syllabification corpus creation and deep learning applications
Teisovi Angami, Kevisino Khate
https://arxiv.org/abs/2510.00629 https://arxiv.org/p…

Tenyidie Syllabification corpus creation and deep learning applications
The Tenyidie language is a low-resource language of the Tibeto-Burman family spoken by the Tenyimia Community of Nagaland in the north-eastern part of India and is considered a major language in Nagaland. It is tonal, Subject-Object-Verb, and highly agglutinative in nature. Being a low-resource language, very limited research on Natural Language Processing (NLP) has been conducted. To the best of our knowledge, no work on syllabification has been reported for this language. Among the many NLP t…

STARE: Predicting Decision Making Based on Spatio-Temporal Eye Movements
Moshe Unger, Alexander Tuzhilin, Michel Wedel
https://arxiv.org/abs/2508.04148 https://

STARE: Predicting Decision Making Based on Spatio-Temporal Eye Movements
The present work proposes a Deep Learning architecture for the prediction of various consumer choice behaviors from time series of raw gaze or eye fixations on images of the decision environment, for which currently no foundational models are available. The architecture, called STARE (Spatio-Temporal Attention Representation for Eye Tracking), uses a new tokenization strategy, which involves mapping the x- and y- pixel coordinates of eye-movement time series on predefined, contiguous Regions of…

Thermoelectric transport in graphene under strain fields modeled by Dirac oscillators
Juan A. Ca\~nas, Daniel A. Bonilla, A. Mart\'in-Ruiz
https://arxiv.org/abs/2509.04704 h…

Thermoelectric transport in graphene under strain fields modeled by Dirac oscillators
Graphene has emerged as a paradigmatic material in condensed matter physics due to its exceptional electronic, mechanical, and thermal properties. A deep understanding of its thermoelectric transport behavior is crucial for the development of novel nanoelectronic and energy-harvesting devices. In this work, we investigate the thermoelectric transport properties of monolayer graphene subjected to randomly distributed localized strain fields, which locally induce impurity-like perturbations. Thes…

OSC: Cognitive Orchestration through Dynamic Knowledge Alignment in Multi-Agent LLM Collaboration
Jusheng Zhang, Yijia Fan, Kaitong Cai, Xiaofei Sun, Keze Wang
https://arxiv.org/abs/2509.04876

OSC: Cognitive Orchestration through Dynamic Knowledge Alignment in Multi-Agent LLM Collaboration
This paper introduces OSC (Orchestrating Cognitive Synergy), a knowledge-aware adaptive collaboration framework designed to enhance cognitive synergy in multi-agent systems with large language models. While prior work has advanced agent selection and result aggregation, efficient linguistic interactions for deep collaboration among expert agents remain a critical bottleneck. OSC addresses this gap as a pivotal intermediate layer between selection and aggregation, introducing Collaborator Knowle…

Representing Beauty: Towards a Participatory but Objective Latent Aesthetics
Alexander Michael Rusnak
https://arxiv.org/abs/2510.02869 https://arxiv.org/pd…

Representing Beauty: Towards a Participatory but Objective Latent Aesthetics
What does it mean for a machine to recognize beauty? While beauty remains a culturally and experientially compelling but philosophically elusive concept, deep learning systems increasingly appear capable of modeling aesthetic judgment. In this paper, we explore the capacity of neural networks to represent beauty despite the immense formal diversity of objects for which the term applies. By drawing on recent work on cross-model representational convergence, we show how aesthetic content produces…

Started work on the PIC12F683 deep dive project.
Here's a WIP shot of the cross section at P2500. Needs just a little more time to center the section in the wire bonds before I switch to finer abrasives for the final polish.

OpenAI debuts ChatGPT Agent, which can control an entire computer and perform multi-step tasks, powered by a new dedicated model, rolling out to paid users (Hayden Field/The Verge)
https://www.theverge.com/ai-artificial-intel…

OpenAI’s new ChatGPT Agent can control an entire computer and do tasks for you
OpenAI debuted ChatGPT Agent, a tool that can complete work on your behalf using its own “virtual computer” and is a continuation of its Operator and Deep Research tools. 

Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks
Jason Gardner, Ayan Dutta, Swapnoneel Roy, O. Patrick Kreidl, Ladislau Boloni
https://arxiv.org/abs/2509.05273

Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks
The growing computational demands of deep reinforcement learning (DRL) have raised concerns about the environmental and economic costs of training large-scale models. While algorithmic efficiency in terms of learning performance has been extensively studied, the energy requirements, greenhouse gas emissions, and monetary costs of DRL algorithms remain largely unexplored. In this work, we present a systematic benchmarking study of the energy consumption of seven state-of-the-art DRL algorithms, …

Error analysis for the deep Kolmogorov method
Iulian C\^impean, Thang Do, Lukas Gonon, Arnulf Jentzen, Ionel Popescu
https://arxiv.org/abs/2508.17167 https://

Error analysis for the deep Kolmogorov method
The deep Kolmogorov method is a simple and popular deep learning based method for approximating solutions of partial differential equations (PDEs) of the Kolmogorov type. In this work we provide an error analysis for the deep Kolmogorov method for heat PDEs. Specifically, we reveal convergence with convergence rates for the overall mean square distance between the exact solution of the heat PDE and the realization function of the approximating deep neural network (DNN) associated with a stochas…

A Modular, Adaptive, and Scalable Quantum Factoring Algorithm
Alok Shukla, Prakash Vedula
https://arxiv.org/abs/2509.05010 https://arxiv.org/pdf/2509.05010…

A Modular, Adaptive, and Scalable Quantum Factoring Algorithm
Shor's algorithm for integer factorization offers an exponential speedup over classical methods but remains impractical on Noisy Intermediate Scale Quantum (NISQ) hardware due to the need for many coherent qubits and very deep circuits. Building on our recent work on adaptive and windowed phase-estimation methods, we have developed a modular, windowed formulation of Shor's algorithm that mitigates these limitations by restructuring phase estimation into shallow, independent circuit blocks that …

Detecting and measuring respiratory events in horses during exercise with a microphone: deep learning vs. standard signal processing
Jeanne I. M. Parmentier (Utrecht University, University of Twente, Inertia Technology B.V), Rhana M. Aarts (Utrecht University), Elin Hernlund (Swedish University of Agricultural Sciences), Marie Rhodin (Swedish University of Agricultural Sciences), Berend Jan van der Zwaag (University of Twente, Inertia Technology B.V)

Detecting and measuring respiratory events in horses during exercise with a microphone: deep learning vs. standard signal processing
Monitoring respiration parameters such as respiratory rate could be beneficial to understand the impact of training on equine health and performance and ultimately improve equine welfare. In this work, we compare deep learning-based methods to an adapted signal processing method to automatically detect cyclic respiratory events and extract the dynamic respiratory rate from microphone recordings during high intensity exercise in Standardbred trotters. Our deep learning models are able to detect …

Is 42 the answer to the ultimate question of life, the universe, and everything? Work on The vOICe sensory substitution device for the blind has been ongoing for 42 years now.
Grok: https://x.com/i/grok/share/FcYnQAJ8SxmnoyisMtCgkxz6X
ChatGPT:

Is 42 the answer to the ultimate question of life, the universe, and everything? Work on The vOICe visual-to-auditory sensory substitution device for the blind has been ongoing for 42 years now, starting in 1983.
Yes, according to Douglas Adams' The Hitchhiker's Guide to the Galaxy, 42 is indeed the answer to the ultimate question of life, the universe, and everything—though the question itself remains a mystery, as the supercomputer Deep Thought spent 7.5 million years calculating it only to leave everyone hanging. As for The vOICe (which cleverly stands for

Cowboys camp observations: Deep ball needs work, pass rushers look good in pads https://www.nytimes.com/athletic/6519279/2025/07/28/cowboys-training-camp-passing-game-dak-prescott/

And here we go, another grant proposal rejected. You better not collaborate with me as I'm quite successful in not getting funding. That's the fifth in a row, if I counted correctly. Basically all work of the last 12 months completely wasted
And as our university is run as a profit-oriented business with controlling and such, I'm knee-deep in the red with hundreds of minus hours which might result in getting paid less. Despite being completely overworked

Super-Penrose $\And$ Witten Transforms for SCFT$_3$
Deep Mazumdar
https://arxiv.org/abs/2508.02672 https://arxiv.org/pdf/2508.02672

Super-Penrose $\And$ Witten Transforms for SCFT$_3$
The study of three dimensional CFT correlators in twistor space has recently garnered a significant interest. Conformal symmetry acts linearly in the twistor space, which streamlines the analysis. Moreover, twistors provide a connection to the position and momentum space through the Penrose and Witten transforms, respectively. In this work, we develop the supersymmetric versions of Penrose and Witten transforms for three dimensional superconformal field theories for $\mathcal{N}=1\;\textrm{to}\…

Automated Type Annotation in Python Using Large Language Models
Varun Bharti, Shashwat Jha, Dhruv Kumar, Pankaj Jalote
https://arxiv.org/abs/2508.00422 https://

Automated Type Annotation in Python Using Large Language Models
Type annotations in Python enhance maintainability and error detection. However, generating these annotations manually is error prone and requires extra effort. Traditional automation approaches like static analysis, machine learning, and deep learning struggle with limited type vocabularies, behavioral over approximation, and reliance on large labeled datasets. In this work, we explore the use of LLMs for generating type annotations in Python. We develop a generate check repair pipeline: the L…

Revisiting MFCCs: Evidence for Spectral-Prosodic Coupling
Vitor Magno de O. S. Bezerra, Gabriel F. A. Bastos, Jugurta Montalv\~ao
https://arxiv.org/abs/2510.05922 https://

Revisiting MFCCs: Evidence for Spectral-Prosodic Coupling
Mel-frequency cepstral coefficients (MFCCs) are an important feature in speech processing. A deeper understanding of their properties can contribute to the work that is being done with both classical and deep learning models. This study challenges the long-held assumption that MFCCs lack relevant temporal information by investigating their relationship with speech prosody. Using a null hypothesis significance testing framework, a systematic assessment is made about the statistical independence …

Defending against Stegomalware in Deep Neural Networks with Permutation Symmetry
Birk Torpmann-Hagen, Michael A. Riegler, P{\aa}l Halvorsen, Dag Johansen
https://arxiv.org/abs/2509.20399

Defending against Stegomalware in Deep Neural Networks with Permutation Symmetry
Deep neural networks are being utilized in a growing number of applications, both in production systems and for personal use. Network checkpoints are as a consequence often shared and distributed on various platforms to ease the development process. This work considers the threat of neural network stegomalware, where malware is embedded in neural network checkpoints at a negligible cost to network accuracy. This constitutes a significant security concern, but is nevertheless largely neglected b…

Accurate and scalable deep Maxwell solvers using multilevel iterative methods
Chenkai Mao, Jonathan A. Fan
https://arxiv.org/abs/2509.03622 https://arxiv.o…

Accurate and scalable deep Maxwell solvers using multilevel iterative methods
Neural networks have promise as surrogate partial differential equation (PDE) solvers, but it remains a challenge to use these concepts to solve problems with high accuracy and scalability. In this work, we show that neural network surrogates can combine with iterative algorithms to accurately solve PDE problems featuring different scales, resolutions, and boundary conditions. We develop a subdomain neural operator model that supports arbitrary Robin-type boundary condition inputs, and we show …

CardiacFlow: 3D t Four-Chamber Cardiac Shape Completion and Generation via Flow Matching
Qiang Ma, Qingjie Meng, Mengyun Qiao, Paul M. Matthews, Declan P. O'Regan, Wenjia Bai
https://arxiv.org/abs/2509.05754

CardiacFlow: 3D+t Four-Chamber Cardiac Shape Completion and Generation via Flow Matching
Learning 3D+t shape completion and generation from multi-view cardiac magnetic resonance (CMR) images requires a large amount of high-resolution 3D whole-heart segmentations (WHS) to capture shape priors. In this work, we leverage flow matching techniques to learn deep generative flows for augmentation, completion, and generation of 3D+t shapes of four cardiac chambers represented implicitly by segmentations. Firstly, we introduce a latent rectified flow to generate 3D cardiac shapes for data a…

Assessment of deep learning models integrated with weather and environmental variables for wildfire spread prediction and a case study of the 2023 Maui fires
Jiyeon Kim, Yingjie Hu, Negar Elhami-Khorasani, Kai Sun, Ryan Zhenqi Zhou
https://arxiv.org/abs/2509.21327

Assessment of deep learning models integrated with weather and environmental variables for wildfire spread prediction and a case study of the 2023 Maui fires
Predicting the spread of wildfires is essential for effective fire management and risk assessment. With the fast advancements of artificial intelligence (AI), various deep learning models have been developed and utilized for wildfire spread prediction. However, there is limited understanding of the advantages and limitations of these models, and it is also unclear how deep learning-based fire spread models can be compared with existing non-AI fire models. In this work, we assess the ability of …

MultiSurv: A Multimodal Deep Survival Framework for Prostrate and Bladder Cancer
Noorul Wahab, Ethar Alzaid, Jiaqi Lv, Adam Shephard, Shan E Ahmed Raza
https://arxiv.org/abs/2509.05037

MultiSurv: A Multimodal Deep Survival Framework for Prostrate and Bladder Cancer
Accurate prediction of time-to-event outcomes is a central challenge in oncology, with significant implications for treatment planning and patient management. In this work, we present MultiSurv, a multimodal deep survival model utilising DeepHit with a projection layer and inter-modality cross-attention, which integrates heterogeneous patient data, including clinical, MRI, RNA-seq and whole-slide pathology features. The model is designed to capture complementary prognostic signals across modali…

I often find it gut-wrenching to watch the snail’s pace at which established scientific knowledge trickles into the public consciousness.
The study of income inequality is a good example.
In 2001, Thomas Piketty and Emmanuel Saez published a landmark paper which showed, for the first time, that US top income shares had been on the rise since the 1980s.
Piketty and Saez’s work has since been replicated and expanded numerous times.
In short, scientists know that the US …

Partisan Politics and the Road to Plutocracy – Economics from the Top Down
In this essay, I take a deep dive into the history of American plutocracy and its conspicuously Republican roots.

Label-frugal satellite image change detection with generative virtual exemplar learning
Hichem Sahbi
https://arxiv.org/abs/2510.06926 https://arxiv.org/pdf…

Label-frugal satellite image change detection with generative virtual exemplar learning
Change detection is a major task in remote sensing which consists in finding all the occurrences of changes in multi-temporal satellite or aerial images. The success of existing methods, and particularly deep learning ones, is tributary to the availability of hand-labeled training data that capture the acquisition conditions and the subjectivity of the user (oracle). In this paper, we devise a novel change detection algorithm, based on active learning. The main contribution of our work resides …

Should we teach vibe coding? Here's why not.
Should AI coding be taught in undergrad CS education?
1/2
I teach undergraduate computer science labs, including for intro and more-advanced core courses. I don't publish (non-negligible) scholarly work in the area, but I've got years of craft expertise in course design, and I do follow the academic literature to some degree. In other words, In not the world's leading expert, but I have spent a lot of time thinking about course design, and consider myself competent at it, with plenty of direct experience in what knowledge & skills I can expect from students as they move through the curriculum.
I'm also strongly against most uses of what's called "AI" these days (specifically, generative deep neutral networks as supplied by our current cadre of techbro). There are a surprising number of completely orthogonal reasons to oppose the use of these systems, and a very limited number of reasonable exceptions (overcoming accessibility barriers is an example). On the grounds of environmental and digital-commons-pollution costs alone, using specifically the largest/newest models is unethical in most cases.
But as any good teacher should, I constantly question these evaluations, because I worry about the impact on my students should I eschew teaching relevant tech for bad reasons (and even for his reasons). I also want to make my reasoning clear to students, who should absolutely question me on this. That inspired me to ask a simple question: ignoring for one moment the ethical objections (which we shouldn't, of course; they're very stark), at what level in the CS major could I expect to teach a course about programming with AI assistance, and expect students to succeed at a more technically demanding final project than a course at the same level where students were banned from using AI? In other words, at what level would I expect students to actually benefit from AI coding "assistance?"
To be clear, I'm assuming that students aren't using AI in other aspects of coursework: the topic of using AI to "help you study" is a separate one (TL;DR it's gross value is not negative, but it's mostly not worth the harm to your metacognitive abilities, which AI-induced changes to the digital commons are making more important than ever).
So what's my answer to this question?
If I'm being incredibly optimistic, senior year. Slightly less optimistic, second year of a masters program. Realistic? Maybe never.
The interesting bit for you-the-reader is: why is this my answer? (Especially given that students would probably self-report significant gains at lower levels.) To start with, [this paper where experienced developers thought that AI assistance sped up their work on real tasks when in fact it slowed it down] (https://arxiv.org/abs/2507.09089) is informative. There are a lot of differences in task between experienced devs solving real bugs and students working on a class project, but it's important to understand that we shouldn't have a baseline expectation that AI coding "assistants" will speed things up in the best of circumstances, and we shouldn't trust self-reports of productivity (or the AI hype machine in general).
Now we might imagine that coding assistants will be better at helping with a student project than at helping with fixing bugs in open-source software, since it's a much easier task. For many programming assignments that have a fixed answer, we know that many AI assistants can just spit out a solution based on prompting them with the problem description (there's another elephant in the room here to do with learning outcomes regardless of project success, but we'll ignore this over too, my focus here is on project complexity reach, not learning outcomes). My question is about more open-ended projects, not assignments with an expected answer. Here's a second study (by one of my colleagues) about novices using AI assistance for programming tasks. It showcases how difficult it is to use AI tools well, and some of these stumbling blocks that novices in particular face.
But what about intermediate students? Might there be some level where the AI is helpful because the task is still relatively simple and the students are good enough to handle it? The problem with this is that as task complexity increases, so does the likelihood of the AI generating (or copying) code that uses more complex constructs which a student doesn't understand. Let's say I have second year students writing interactive websites with JavaScript. Without a lot of care that those students don't know how to deploy, the AI is likely to suggest code that depends on several different frameworks, from React to JQuery, without actually setting up or including those frameworks, and of course three students would be way out of their depth trying to do that. This is a general problem: each programming class carefully limits the specific code frameworks and constructs it expects students to know based on the material it covers. There is no feasible way to limit an AI assistant to a fixed set of constructs or frameworks, using current designs. There are alternate designs where this would be possible (like AI search through adaptation from a controlled library of snippets) but those would be entirely different tools.
So what happens on a sizeable class project where the AI has dropped in buggy code, especially if it uses code constructs the students don't understand? Best case, they understand that they don't understand and re-prompt, or ask for help from an instructor or TA quickly who helps them get rid of the stuff they don't understand and re-prompt or manually add stuff they do. Average case: they waste several hours and/or sweep the bugs partly under the rug, resulting in a project with significant defects. Students in their second and even third years of a CS major still have a lot to learn about debugging, and usually have significant gaps in their knowledge of even their most comfortable programming language. I do think regardless of AI we as teachers need to get better at teaching debugging skills, but the knowledge gaps are inevitable because there's just too much to know. In Python, for example, the LLM is going to spit out yields, async functions, try/finally, maybe even something like a while/else, or with recent training data, the walrus operator. I can't expect even a fraction of 3rd year students who have worked with Python since their first year to know about all these things, and based on how students approach projects where they have studied all the relevant constructs but have forgotten some, I'm not optimistic seeing these things will magically become learning opportunities. Student projects are better off working with a limited subset of full programming languages that the students have actually learned, and using AI coding assistants as currently designed makes this impossible. Beyond that, even when the "assistant" just introduces bugs using syntax the students understand, even through their 4th year many students struggle to understand the operation of moderately complex code they've written themselves, let alone written by someone else. Having access to an AI that will confidently offer incorrect explanations for bugs will make this worse.
To be sure a small minority of students will be able to overcome these problems, but that minority is the group that has a good grasp of the fundamentals and has broadened their knowledge through self-study, which earlier AI-reliant classes would make less likely to happen. In any case, I care about the average student, since we already have plenty of stuff about our institutions that makes life easier for a favored few while being worse for the average student (note that our construction of that favored few as the "good" students is a large part of this problem).
To summarize: because AI assistants introduce excess code complexity and difficult-to-debug bugs, they'll slow down rather than speed up project progress for the average student on moderately complex projects. On a fixed deadline, they'll result in worse projects, or necessitate less ambitious project scoping to ensure adequate completion, and I expect this remains broadly true through 4-6 years of study in most programs (don't take this as an endorsement of AI "assistants" for masters students; we've ignored a lot of other problems along the way).
There's a related problem: solving open-ended project assignments well ultimately depends on deeply understanding the problem, and AI "assistants" allow students to put a lot of code in their file without spending much time thinking about the problem or building an understanding of it. This is awful for learning outcomes, but also bad for project success. Getting students to see the value of thinking deeply about a problem is a thorny pedagogical puzzle at the best of times, and allowing the use of AI "assistants" makes the problem much much worse. This is another area I hope to see (or even drive) pedagogical improvement in, for what it's worth.
1/2

BACHI: Boundary-Aware Symbolic Chord Recognition Through Masked Iterative Decoding on Pop and Classical Music
Mingyang Yao, Ke Chen, Shlomo Dubnov, Taylor Berg-Kirkpatrick
https://arxiv.org/abs/2510.06528

BACHI: Boundary-Aware Symbolic Chord Recognition Through Masked Iterative Decoding on Pop and Classical Music
Automatic chord recognition (ACR) via deep learning models has gradually achieved promising recognition accuracy, yet two key challenges remain. First, prior work has primarily focused on audio-domain ACR, while symbolic music (e.g., score) ACR has received limited attention due to data scarcity. Second, existing methods still overlook strategies that are aligned with human music analytical practices. To address these challenges, we make two contributions: (1) we introduce POP909-CL, an enhance…

Deep learning for jet modification in the presence of the QGP background
Ran Li, Yi-Lun Du, Shanshan Cao
https://arxiv.org/abs/2508.20856 https://arxiv.org…

Deep learning for jet modification in the presence of the QGP background
Jet interactions with the color-deconfined QCD medium in relativistic heavy-ion collisions are conventionally assessed by measuring the modification of the distributions of jet observables with respect to their baselines in proton-proton collisions. Deep learning methods enable per-jet evaluation of these modifications, enhancing the use of jets as precision probes of the nuclear medium. In this work, we predict the jet-by-jet fractional energy loss $χ$ for jets evolving through a quark-gluon …

Barlow-Swin: Toward a novel siamese-based segmentation architecture using Swin-Transformers
Morteza Kiani Haftlang, Mohammadhossein Malmir, Foroutan Parand, Umberto Michelucci, Safouane El Ghazouali
https://arxiv.org/abs/2509.06885

Barlow-Swin: Toward a novel siamese-based segmentation architecture using Swin-Transformers
Medical image segmentation is a critical task in clinical workflows, particularly for the detection and delineation of pathological regions. While convolutional architectures like U-Net have become standard for such tasks, their limited receptive field restricts global context modeling. Recent efforts integrating transformers have addressed this, but often result in deep, computationally expensive models unsuitable for real-time use. In this work, we present a novel end-to-end lightweight archi…

ShapeGen3DCP: A Deep Learning Framework for Layer Shape Prediction in 3D Concrete Printing
Giacomo Rizzieri, Federico Lanteri, Liberato Ferrara, Massimiliano Cremonesi
https://arxiv.org/abs/2510.02009 …

ShapeGen3DCP: A Deep Learning Framework for Layer Shape Prediction in 3D Concrete Printing
This work introduces ShapeGen3DCP, a deep learning framework for fast and accurate prediction of filament cross-sectional geometry in 3D Concrete Printing (3DCP). The method is based on a neural network architecture that takes as input both material properties in the fluid state (density, yield stress, plastic viscosity) and process parameters (nozzle diameter, nozzle height, printing and flow velocities) to directly predict extruded layer shapes. To enhance generalization, some inputs are refo…

Deep Reinforcement Learning for Ranking Utility Tuning in the Ad Recommender System at Pinterest
Xiao Yang, Mehdi Ben Ayed, Longyu Zhao, Fan Zhou, Yuchen Shen, Abe Engle, Jinfeng Zhuang, Ling Leng, Jiajing Xu, Charles Rosenberg, Prathibha Deshikachar
https://arxiv.org/abs/2509.05292

Deep Reinforcement Learning for Ranking Utility Tuning in the Ad Recommender System at Pinterest
The ranking utility function in an ad recommender system, which linearly combines predictions of various business goals, plays a central role in balancing values across the platform, advertisers, and users. Traditional manual tuning, while offering simplicity and interpretability, often yields suboptimal results due to its unprincipled tuning objectives, the vast amount of parameter combinations, and its lack of personalization and adaptability to seasonality. In this work, we propose a general…

Benchmarking Deep Learning Convolutions on Energy-constrained CPUs
Enrique Galvez (ALSOC), Adrien Cassagne (ALSOC), Alix Munier (ALSOC), Manuel Bouyer
https://arxiv.org/abs/2509.26217

Benchmarking Deep Learning Convolutions on Energy-constrained CPUs
This work evaluates state-of-the-art convolution algorithms for CPU-based deep learning inference. While most prior studies focus on GPUs or NPUs, CPU implementations remain relatively underoptimized. We benchmark direct, GEMM-based, and Winograd convolutions across modern CPUs from ARM __ , Intel __ , AMD __ , Apple __ , and Nvidia __ , considering both latency and energy efficiency. Our results highlight the key architectural factors that govern CPU efficiency for convolution operations, prov…

The Clustering of Active Galactic Nuclei and Star Forming Galaxies in the LoTSS DeepFields
C. L. Hale, P. N. Best, K. J. Duncan, R. Kondapally, M. J. Jarvis, M. Magliocchetti, H. J. A. R\"ottgering, D. J. Schwarz, D. J. B. Smith, J. Zheng
https://arxiv.org/abs/2510.01029

The Clustering of Active Galactic Nuclei and Star Forming Galaxies in the LoTSS DeepFields
Using deep observations across three of the LOFAR Two-metre Sky Survey Deep Fields, this work measures the angular clustering of star forming galaxies (SFGs) and low-excitation radio galaxies (LERGs) to $z$$\lesssim$1.5 for faint sources, $S_{\textrm{144 MHz}}$$\geq$200 $μ$Jy. We measure the angular auto-correlation of LOFAR sources in redshift bins and their cross-correlation with multi-wavelength sources {to} measure the evolving galaxy bias for SFGs and LERGs. Our work shows the bias of the…

A Comprehensive Framework for Uncertainty Quantification of Voxel-wise Supervised Models in IVIM MRI
Nicola Casali, Alessandro Brusaferri, Giuseppe Baselli, Stefano Fumagalli, Edoardo Micotti, Gianluigi Forloni, Riaz Hussein, Giovanna Rizzo, Alfonso Mastropietro
https://arxiv.org/abs/2508.04588

A Comprehensive Framework for Uncertainty Quantification of Voxel-wise Supervised Models in IVIM MRI
Accurate estimation of intravoxel incoherent motion (IVIM) parameters from diffusion-weighted MRI remains challenging due to the ill-posed nature of the inverse problem and high sensitivity to noise, particularly in the perfusion compartment. In this work, we propose a probabilistic deep learning framework based on Deep Ensembles (DE) of Mixture Density Networks (MDNs), enabling estimation of total predictive uncertainty and decomposition into aleatoric (AU) and epistemic (EU) components. The m…

JADES Data Release 4 -- Paper II: Data reduction, analysis and emission-line fluxes of the complete spectroscopic sample
J. Scholtz, S. Carniani, E. Parlanti, F. D'Eugenio, E. Curtis-Lake, P. Jakobsen, A. J. Bunker, A. J. Cameron, S. Arribas, W. M. Baker, S. Charlot, J. Chevellard, C. Circosta, M. Curti, Q. Duan, D. J. Eisenstein, K. Hainline, Z. Ji, B. D. Johnson, G. C. Jones, N. Kumari, R. Maiolino, M. V. Maseda, M. Perna, P. G. P\'erez-Gonz\'alez, T. Rawle, M. Rieke, P. …

JADES Data Release 4 - Paper II: Data reduction, analysis and emission-line fluxes of the complete spectroscopic sample
We present the fourth data release of JADES, the JWST Advanced Deep Extragalactic Survey, providing deep spectroscopic observations in the two GOODS fields. A companion paper presents the target selection, spectroscopic redshifts and success rates, and in this paper, we discuss the data reduction and present emission line flux measurements. The spectroscopy in this work consists of medium-depth, deep and ultradeep NIRSpec/MSA spectra of 5,190 targets, covering the spectral range $0.6\text{--}5.…

Deep Learning for CMB Foreground Removal and Beam Deconvolution: A U-Net GAN Approach
Obasho M, Shambhavi Jaiswal, Santanu Das, Krishna Mohan Parattu
https://arxiv.org/abs/2509.00139

Deep Learning for CMB Foreground Removal and Beam Deconvolution: A U-Net GAN Approach
Extracting cosmological information from microwave sky observations requires accurate estimation of the underlying Cosmic Microwave Background (CMB) by removing foreground contamination, instrumental noise, and the effects of beam convolution. In this work, we develop a machine learning-based approach for CMB reconstruction using a generative adversarial network (GAN) architecture, where the generator is modeled as a U-Net-based convolutional neural network. To train the network, we generate …

Deep Kernel Bayesian Optimisation for Closed-Loop Electrode Microstructure Design with User-Defined Properties based on GANs
Andrea Gayon-Lombardo, Ehecatl A. del Rio-Chanona, Catalina A. Pino-Munoz, Nigel P. Brandon
https://arxiv.org/abs/2508.00833

Deep Kernel Bayesian Optimisation for Closed-Loop Electrode Microstructure Design with User-Defined Properties based on GANs
The generation of multiphase porous electrode microstructures with optimum morphological and transport properties is essential in the design of improved electrochemical energy storage devices, such as lithium-ion batteries. Electrode characteristics directly influence battery performance by acting as the main sites where the electrochemical reactions coupled with transport processes occur. This work presents a generation-optimisation closed-loop algorithm for the design of microstructures with …

Generalizing Unsupervised Lidar Odometry Model from Normal to Snowy Weather Conditions
Beibei Zhou, Zhiyuan Zhang, Zhenbo Song, Jianhui Guo, Hui Kong
https://arxiv.org/abs/2509.02011

Generalizing Unsupervised Lidar Odometry Model from Normal to Snowy Weather Conditions
Deep learning-based LiDAR odometry is crucial for autonomous driving and robotic navigation, yet its performance under adverse weather, especially snowfall, remains challenging. Existing models struggle to generalize across conditions due to sensitivity to snow-induced noise, limiting real-world use. In this work, we present an unsupervised LiDAR odometry model to close the gap between clear and snowy weather conditions. Our approach focuses on effective denoising to mitigate the impact of snow…

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem
Mingfeng Fan, Jiaqi Cheng, Yaoxin Wu, Yifeng Zhang, Yibin Yang, Guohua Wu, Guillaume Sartoretti
https://arxiv.org/abs/2510.03065

A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem
In recent years, deep reinforcement learning (DRL) has gained traction for solving the NP-hard traveling salesman problem (TSP). However, limited attention has been given to the close-enough TSP (CETSP), primarily due to the challenge introduced by its neighborhood-based visitation criterion, wherein a node is considered visited if the agent enters a compact neighborhood around it. In this work, we formulate a Markov decision process (MDP) for CETSP using a discretization scheme and propose a n…

Frugal, Flexible, Faithful: Causal Data Simulation via Frengression
Linying Yang, Robin J. Evans, Xinwei Shen
https://arxiv.org/abs/2508.01018 https://arxi…

Frugal, Flexible, Faithful: Causal Data Simulation via Frengression
Machine learning has revitalized causal inference by combining flexible models and principled estimators, yet robust benchmarking and evaluation remain challenging with real-world data. In this work, we introduce frengression, a deep generative realization of the frugal parameterization that models the joint distribution of covariates, treatments and outcomes around the causal margin of interest. Frengression provides accurate estimation and flexible, faithful simulation of multivariate, time-v…

From Sublinear to Linear: Fast Convergence in Deep Networks via Locally Polyak-Lojasiewicz Regions
Agnideep Aich, Ashit Baran Aich, Bruce Wade
https://arxiv.org/abs/2507.21429 h…

From Sublinear to Linear: Fast Convergence in Deep Networks via Locally Polyak-Lojasiewicz Regions
The convergence of gradient descent (GD) on the non-convex loss landscapes of deep neural networks (DNNs) presents a fundamental theoretical challenge. While recent work has established that GD converges to a stationary point at a sublinear rate within locally quasi-convex regions (LQCRs), this fails to explain the exponential convergence rates consistently observed in practice. In this paper, we resolve this discrepancy by proving that under a mild assumption on Neural Tangent Kernel (NTK) sta…

Forensic Similarity for Speech Deepfakes
Viola Negroni, Davide Salvi, Daniele Ugo Leonzio, Paolo Bestagini, Stefano Tubaro
https://arxiv.org/abs/2510.02864 https://

Forensic Similarity for Speech Deepfakes
In this paper, we introduce a digital audio forensics approach called Forensic Similarity for Speech Deepfakes, which determines whether two audio segments contain the same forensic traces or not. Our work is inspired by prior work in the image domain on forensic similarity, which proved strong generalization capabilities against unknown forensic traces, without requiring prior knowledge of them at training time. To achieve this in the audio setting, we propose a two-part deep-learning system c…

Day 6: Kamome Shirahama
Before I wander much father afield, I'd be remiss not to include at least one Mangaka (I've got 8 on my planning list; if you think Manga is pushing it just wait until you see what the next few days have in store).
I'm currently following "Witch Hat Atelier," and it's absolutely amazing in several dimensions: first class world-building, deep philosophical themes, nuanced diverse cast, tightly-constructed interwoven plots, deep mysteries that keep everything churning and show up in unexpected places, absolutely stellar art both in terms of in-panel depictions and page layouts (some are Watchmen-quality), especially if you are sartorially inclined, and general kindness of its core messages. This is a series I wish every programmer would read, because it includes excellent advice about software design in multiple ways (did I mention there's an intricate and logical magic system within which the main character innovates in legible-to-the-reader-as-innovation ways?). Also, I bet I would have enjoyed this just a much as a 10-year-old as I'm enjoying it in my 30's, which is something that takes well-honed skill to pull off.
Shirahama is a master of her craft, and I'm honestly kinda surprised to see Witch Hat is only her second series. Definitely thinking how I can get my hands on her earlier work in English.
#20AuthorsNoMen

CapsBeam: Accelerating Capsule Network based Beamformer for Ultrasound Non-Steered Plane Wave Imaging on Field Programmable Gate Array
Abdul Rahoof, Vivek Chaturvedi, Mahesh Raveendranatha Panicker, Muhammad Shafique
https://arxiv.org/abs/2509.03201

CapsBeam: Accelerating Capsule Network based Beamformer for Ultrasound Non-Steered Plane Wave Imaging on Field Programmable Gate Array
In recent years, there has been a growing trend in accelerating computationally complex non-real-time beamforming algorithms in ultrasound imaging using deep learning models. However, due to the large size and complexity these state-of-the-art deep learning techniques poses significant challenges when deploying on resource-constrained edge devices. In this work, we propose a novel capsule network based beamformer called CapsBeam, designed to operate on raw radio-frequency data and provide an en…

Beyond Random: Automatic Inner-loop Optimization in Dataset Distillation
Muquan Li, Hang Gou, Dongyang Zhang, Shuang Liang, Xiurui Xie, Deqiang Ouyang, Ke Qin
https://arxiv.org/abs/2510.04838

Beyond Random: Automatic Inner-loop Optimization in Dataset Distillation
The growing demand for efficient deep learning has positioned dataset distillation as a pivotal technique for compressing training dataset while preserving model performance. However, existing inner-loop optimization methods for dataset distillation typically rely on random truncation strategies, which lack flexibility and often yield suboptimal results. In this work, we observe that neural networks exhibit distinct learning dynamics across different training stages-early, middle, and late-maki…

An Efficient Subspace Algorithm for Federated Learning on Heterogeneous Data
Jiaojiao Zhang, Yuqi Xu, Kun Yuan
https://arxiv.org/abs/2509.05213 https://arx…

An Efficient Subspace Algorithm for Federated Learning on Heterogeneous Data
This work addresses the key challenges of applying federated learning to large-scale deep neural networks, particularly the issue of client drift due to data heterogeneity across clients and the high costs of communication, computation, and memory. We propose FedSub, an efficient subspace algorithm for federated learning on heterogeneous data. Specifically, FedSub utilizes subspace projection to guarantee local updates of each client within low-dimensional subspaces, thereby reducing communicat…

Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
Xin Gu, Yingtai Xiao, Guanlin He, Jiamu Bai, Daniel Kifer, Kiwan Maeng
https://arxiv.org/abs/2510.05416 https:…

Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
Differentially private stochastic gradient descent (DP-SGD) offers the promise of training deep learning models while mitigating many privacy risks. However, there is currently a large accuracy gap between DP-SGD and normal SGD training. This has resulted in different lines of research investigating orthogonal ways of improving privacy-preserving training. One such line of work, known as DP-MF, correlates the privacy noise across different iterations of stochastic gradient descent -- allowing l…

Wave-GMS: Lightweight Multi-Scale Generative Model for Medical Image Segmentation
Talha Ahmed, Nehal Ahmed Shaikh, Hassan Mohy-ud-Din
https://arxiv.org/abs/2510.03216 https://…

Wave-GMS: Lightweight Multi-Scale Generative Model for Medical Image Segmentation
For equitable deployment of AI tools in hospitals and healthcare facilities, we need Deep Segmentation Networks that offer high performance and can be trained on cost-effective GPUs with limited memory and large batch sizes. In this work, we propose Wave-GMS, a lightweight and efficient multi-scale generative model for medical image segmentation. Wave-GMS has a substantially smaller number of trainable parameters, does not require loading memory-intensive pretrained vision foundation models, an…

Improving and Evaluating Open Deep Research Agents
Doaa Allabadi, Kyle Bradbury, Jordan M. Malof
https://arxiv.org/abs/2508.10152 https://arxiv.org/pdf/250…

Improving and Evaluating Open Deep Research Agents
We focus here on Deep Research Agents (DRAs), which are systems that can take a natural language prompt from a user, and then autonomously search for, and utilize, internet-based content to address the prompt. Recent DRAs have demonstrated impressive capabilities on public benchmarks however, recent research largely involves proprietary closed-source systems. At the time of this work, we only found one open-source DRA, termed Open Deep Research (ODR). In this work we adapt the challenging recen…

Linear RNNs for autoregressive generation of long music samples
Konrad Szewczyk, Daniel Gallo Fern\'andez, James Townsend
https://arxiv.org/abs/2510.02401 https://

Linear RNNs for autoregressive generation of long music samples
Directly learning to generate audio waveforms in an autoregressive manner is a challenging task, due to the length of the raw sequences and the existence of important structure on many different timescales. Traditional approaches based on recurrent neural networks, as well as causal convolutions and self-attention, have only had limited success on this task. However, recent work has shown that deep state space models, also referred to as linear RNNs, can be highly efficient in this context. In …

Hybrid Deep Learning and Handcrafted Feature Fusion for Mammographic Breast Cancer Classification
Maximilian Tschuchnig, Michael Gadermayr, Khalifa Djemal
https://arxiv.org/abs/2507.19843

Hybrid Deep Learning and Handcrafted Feature Fusion for Mammographic Breast Cancer Classification
Automated breast cancer classification from mammography remains a significant challenge due to subtle distinctions between benign and malignant tissue. In this work, we present a hybrid framework combining deep convolutional features from a ResNet-50 backbone with handcrafted descriptors and transformer-based embeddings. Using the CBIS-DDSM dataset, we benchmark our ResNet-50 baseline (AUC: 78.1%) and demonstrate that fusing handcrafted features with deep ResNet-50 and DINOv2 features improves …

SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing
Jakub Straka, Ivan Gruber
https://arxiv.org/abs/2508.21402 https://arxiv.org/pdf/2…

SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing
Self-supervised learning has emerged as a powerful tool for remote sensing, where large amounts of unlabeled data are available. In this work, we investigate the use of DINO, a contrastive self-supervised method, for pretraining on remote sensing imagery. We introduce SatDINO, a model tailored for representation learning in satellite imagery. Through extensive experiments on multiple datasets in multiple testing setups, we demonstrate that SatDINO outperforms other state-of-the-art methods base…

Enhancing Diabetic Retinopathy Classification Accuracy through Dual Attention Mechanism in Deep Learning
Abdul Hannan, Zahid Mahmood, Rizwan Qureshi, Hazrat Ali
https://arxiv.org/abs/2507.19199

Enhancing Diabetic Retinopathy Classification Accuracy through Dual Attention Mechanism in Deep Learning
Automatic classification of Diabetic Retinopathy (DR) can assist ophthalmologists in devising personalized treatment plans, making it a critical component of clinical practice. However, imbalanced data distribution in the dataset becomes a bottleneck in the generalization of deep learning models trained for DR classification. In this work, we combine global attention block (GAB) and category attention block (CAB) into the deep learning model, thus effectively overcoming the imbalanced data dist…

Priors Matter: Addressing Misspecification in Bayesian Deep Q-Learning
Pascal R. van der Vaart, Neil Yorke-Smith, Matthijs T. J. Spaan
https://arxiv.org/abs/2508.21488 https://

Priors Matter: Addressing Misspecification in Bayesian Deep Q-Learning
Uncertainty quantification in reinforcement learning can greatly improve exploration and robustness. Approximate Bayesian approaches have recently been popularized to quantify uncertainty in model-free algorithms. However, so far the focus has been on improving the accuracy of the posterior approximation, instead of studying the accuracy of the prior and likelihood assumptions underlying the posterior. In this work, we demonstrate that there is a cold posterior effect in Bayesian deep Q-learnin…

Why Do We Need Warm-up? A Theoretical Perspective
Foivos Alimisis, Rustem Islamov, Aurelien Lucchi
https://arxiv.org/abs/2510.03164 https://arxiv.org/pdf/2…

Why Do We Need Warm-up? A Theoretical Perspective
Learning rate warm-up - increasing the learning rate at the beginning of training - has become a ubiquitous heuristic in modern deep learning, yet its theoretical foundations remain poorly understood. In this work, we provide a principled explanation for why warm-up improves training. We rely on a generalization of the $(L_0, L_1)$-smoothness condition, which bounds local curvature as a linear function of the loss sub-optimality and exhibits desirable closure properties. We demonstrate both the…

Pure-Pass: Fine-Grained, Adaptive Masking for Dynamic Token-Mixing Routing in Lightweight Image Super-Resolution
Junyu Wu, Jie Tang, Jie Liu, Gangshan Wu
https://arxiv.org/abs/2510.01997

Pure-Pass: Fine-Grained, Adaptive Masking for Dynamic Token-Mixing Routing in Lightweight Image Super-Resolution
Image Super-Resolution (SR) aims to reconstruct high-resolution images from low-resolution counterparts, but the computational complexity of deep learning-based methods often hinders practical deployment. CAMixer is the pioneering work to integrate the advantages of existing lightweight SR methods and proposes a content-aware mixer to route token mixers of varied complexities according to the difficulty of content recovery. However, several limitations remain, such as poor adaptability, coarse-…

XQC: Well-conditioned Optimization Accelerates Deep Reinforcement Learning
Daniel Palenicek, Florian Vogt, Joe Watson, Ingmar Posner, Jan Peters
https://arxiv.org/abs/2509.25174

XQC: Well-conditioned Optimization Accelerates Deep Reinforcement Learning
Sample efficiency is a central property of effective deep reinforcement learning algorithms. Recent work has improved this through added complexity, such as larger models, exotic network architectures, and more complex algorithms, which are typically motivated purely by empirical performance. We take a more principled approach by focusing on the optimization landscape of the critic network. Using the eigenspectrum and condition number of the critic's Hessian, we systematically investigate the i…

Superposition disentanglement of neural representations reveals hidden alignment
Andr\'e Longon, David Klindt, Meenakshi Khosla
https://arxiv.org/abs/2510.03186 https://

Superposition disentanglement of neural representations reveals hidden alignment
The superposition hypothesis states that a single neuron within a population may participate in the representation of multiple features in order for the population to represent more features than the number of neurons. In neuroscience and AI, representational alignment metrics measure the extent to which different deep neural networks (DNNs) or brains represent similar information. In this work, we explore a critical question: \textit{does superposition interact with alignment metrics in any un…

Instant4D: 4D Gaussian Splatting in Minutes
Zhanpeng Luo, Haoxi Ran, Li Lu
https://arxiv.org/abs/2510.01119 https://arxiv.org/pdf/2510.01119

Instant4D: 4D Gaussian Splatting in Minutes
Dynamic view synthesis has seen significant advances, yet reconstructing scenes from uncalibrated, casual video remains challenging due to slow optimization and complex parameter estimation. In this work, we present Instant4D, a monocular reconstruction system that leverages native 4D representation to efficiently process casual video sequences within minutes, without calibrated cameras or depth sensors. Our method begins with geometric recovery through deep visual SLAM, followed by grid prunin…

Drop-Muon: Update Less, Converge Faster
Kaja Gruntkowska, Yassine Maziane, Zheng Qu, Peter Richt\'arik
https://arxiv.org/abs/2510.02239 https://arxiv.o…

Drop-Muon: Update Less, Converge Faster
Conventional wisdom in deep learning optimization dictates updating all layers at every step-a principle followed by all recent state-of-the-art optimizers such as Muon. In this work, we challenge this assumption, showing that full-network updates can be fundamentally suboptimal, both in theory and in practice. We introduce a non-Euclidean Randomized Progressive Training method-Drop-Muon-a simple yet powerful framework that updates only a subset of layers per step according to a randomized sche…

Flatness-Aware Stochastic Gradient Langevin Dynamics
Stefano Bruno, Youngsik Hwang, Jaehyeon An, Sotirios Sabanis, Dong-Young Lim
https://arxiv.org/abs/2510.02174 https://

Flatness-Aware Stochastic Gradient Langevin Dynamics
Generalization in deep learning is closely tied to the pursuit of flat minima in the loss landscape, yet classical Stochastic Gradient Langevin Dynamics (SGLD) offers no mechanism to bias its dynamics toward such low-curvature solutions. This work introduces Flatness-Aware Stochastic Gradient Langevin Dynamics (fSGLD), designed to efficiently and provably seek flat minima in high-dimensional nonconvex optimization problems. At each iteration, fSGLD uses the stochastic gradient evaluated at para…

On the Benefits of Weight Normalization for Overparameterized Matrix Sensing
Yudong Wei, Liang Zhang, Bingcong Li, Niao He
https://arxiv.org/abs/2510.01175 https://

On the Benefits of Weight Normalization for Overparameterized Matrix Sensing
While normalization techniques are widely used in deep learning, their theoretical understanding remains relatively limited. In this work, we establish the benefits of (generalized) weight normalization (WN) applied to the overparameterized matrix sensing problem. We prove that WN with Riemannian optimization achieves linear convergence, yielding an exponential speedup over standard methods that do not use WN. Our analysis further demonstrates that both iteration and sample complexity improve p…