#deep learning
82 resultsLarge scale cross-regional remote sensing flood monitoring framework for operative mapping and impact analysis
Ilya Novikov, Svetlana Illarionova, Ruslan Dzharkinov +6
The paper proposes an end‑to‑end multimodal deep‑learning framework that combines SAR, multispectral and elevation data to detect flood water surfaces and assess damage across larg…
Don't Trust the AI Ecosystem: Analyzing Privacy Leakage in Compromised Open-Source Components
Jin-Seong Kim, Han-Ju Lee, Seok-Won Hong +4
The paper presents GradLock, a training-time injection attack that stealthily embeds private training data into model parameters, enabling near‑perfect reconstruction of the data f…
Reliability-calibrated deep residual full-waveform inversion using geometry-invariant physics encoding: synthetic validation and zero-shot Marmousi-2 testing
Deepak Kumar, Jayant Nath Tripathi, Laxmidhar Behera
The paper presents a deep learning pipeline that uses physics‑derived inputs to correct full‑waveform inversion results and provides calibrated, pixel‑wise uncertainty estimates th…
Same Branches, Different Trees: A Bifurcation Connectedness Metric for Coronary Artery Segmentation and FFR-CT Decision Agreement
Maame Owusu-Ansah, Kelvin Lee, Dr Vinod Venugopal +3
The paper introduces a Bifurcation Connectedness Score (BCS) to evaluate how well coronary artery segmentations preserve the connectivity of vessel bifurcations, showing that highe…
Radar-Aided Near-Field Beam Prediction via Beam Map Learning for XL-MIMO V2I Communications
Jiali Nie, Yu Han, Yuanhao Cui +3
The paper introduces a passive radar‑aided framework that predicts near‑field beams for XL‑MIMO vehicle‑to‑infrastructure links by learning a mapping from radar Bartlett spectra to…
Uncertainty quantification for trustworthy deep learning: Methods and measures
H. Martin Gillis, Thomas Trappenberg
The paper surveys methods for quantifying uncertainty in deep neural networks, focusing on ensemble-based and approximate Bayesian approaches and how their outputs are measured.
What to Remove, What to Preserve: Dual-Ambiguity Rectification for All-in-One Image Restoration
Cencen Liu, Wen Yin, Dongyang Zhang +6
The paper introduces DAR-Net, a deep network that tackles the dual ambiguity problem in all‑in‑one image restoration by modeling degradation states with a simplex‑constrained arche…
ARD-REFSM: Enhancing Reflection Symmetry Detection with Asymmetric Denoising and Rotation Equivariance
Dongfu Yin, Rourou Su, Cong Zhao +1
The paper introduces a method that removes asymmetric background clutter and enforces rotation-equivariant feature matching to improve detection of reflection symmetry in images, a…
FootprintNet: State-Transition-Guided Dynamic Footprint Learning for Multi-temporal Remote Sensing Change Detection
Haotian Zhang, Hao Chen, Han Guo +2
The paper introduces FootprintNet, a deep learning framework that detects and classifies dynamic building-change footprints over time in multi-temporal remote sensing images by mod…
From Expert Reduction to Behavioral Divergence: Tracing Numerical State through Sparse MoE Inference
Tianyang Zhu
The paper studies how different expert reduction orders and numerical precision settings cause divergent behavior in sparse mixture-of-experts (MoE) inference, showing that identic…
What Makes Deep Learning Work for Traditional Chinese Medicine Tongue Diagnosis? A Comprehensive Ablation Study
Longxia Gao, Linan Wang, Yuhe Han +3
The paper systematically evaluates how different deep‑learning design choices affect automated tongue diagnosis for traditional Chinese medicine, identifying six key principles for…
Weather Emulators at the Frontier of Heat Extremes Predictability
Cas Decancq, Thomas Mortier, Jessica Keune +1
The paper compares six modern deep‑learning weather emulators with traditional dynamical and statistical models for forecasting global near‑surface temperature and extreme heat at…
PRISM-Net: Patient-specific reference-guided inter-breast symmetry matching for three-class breast DCE-MRI classification
Boya Zhang, Shuaiwen Zhou, Di Kong +7
PRISM-Net is a registration‑free deep learning framework that uses the contralateral breast as a patient‑specific reference to model inter‑breast symmetry and improve classificatio…
CASIAL: Geometric Distortion Robust Image Watermarking
Yupeng Qiu, Han Fang, Ee-Chien Chang
The paper introduces CASIAL, a deep learning framework for image watermarking that spreads watermark bits across the whole image and uses a geometry‑invariant alignment module to s…
An Attention-Based Framework for Alzheimers Disease Classification Using Resting-State fMRI
Harshiddhi Pathak, Gowtham Reddy N, Mrinal Acharya +1
The paper proposes a Transformer‑style attention model that directly processes resting‑state fMRI functional connectivity matrices to classify Alzheimer's disease versus cognitivel…
Surrogate assisted diversity estimation in neural ensemble search
Alexandr Udeneev, Petr Babkin, Oleg Bakhteev
The paper proposes a dual‑objective surrogate‑guided method for neural ensemble search that predicts both accuracy and diversity of candidate architectures, enabling efficient cons…
Skillful forecasting of offshore winds from satellite scatterometer constellations
Francesco Pinto, Luca Lanzilao, Paco Lopez Dekker +1
The paper introduces WindCastNet, a deep‑learning framework that directly uses irregular satellite scatterometer observations to nowcast offshore wind speed and direction, achievin…
Examining the Efficacy of Graph Neural Network Message-Passing in Regression Contexts
Keith G. Mills, Aedan J. DeFrates, Joong Ho Kim
The paper evaluates how different graph neural network message‑passing layers perform on scalar regression tasks, finding that deep convolutional GNNs like GEN generally outperform…
Classification of Disease from Lungs X-ray Images using VGG16, VGG19 and ResNet50 Models
Nand Lal Yadav, Rajesh Kumar, Satyendra Singh +1
The paper evaluates deep convolutional neural networks (VGG16, VGG19, and ResNet50) for classifying lung diseases such as pneumonia, tuberculosis, and lung cancer from chest X‑ray…
Pangram 4 Technical Report
Ben Glickenhaus, Katherine Thai, Jenna Russell +4
The paper introduces Pangram 4, a deep‑learning model for detecting AI‑generated text that achieves high accuracy, strong out‑of‑distribution robustness, and improved detection of…
Emission-Forecasting-Based Spatial-Temporal Carbon Response: A Multi-Agent Attention-Enhanced Deep Learning Framework
Feiyu Cai, Jing Qiu, Yi Yang +4
The paper proposes a deep learning framework with dual-stage attention and a large‑language‑model‑based multi‑agent system to forecast day‑ahead nodal carbon intensity, enabling pr…
Decoding the Micromagnetic Hamiltonian from Magnetic Fingerprints
Bradley J. Fugetta, Anqi Liu, Kai Liu +2
The paper presents deep convolutional neural networks that infer the full micromagnetic Hamiltonian of a material directly from magnetic fingerprint data obtained via First‑Order R…
DO-CGI: deep-optimized illumination patterns for computational ghost imaging at low sampling ratios
Mor Hale, Ofir Lindenbaum, Eliahu Cohen
The paper introduces a deep‑learning framework that designs optimized illumination patterns for computational ghost imaging, achieving higher image quality at very low sampling rat…
Comparing the Performance of Foundation Model Derived Embeddings with Traditional Approaches for Distant Metastasis Prediction in Head and Neck Cancer
Erich Schmitz, Meixu Chen, Bowen Jing +1
The study evaluates CT foundation model embeddings for predicting distant metastasis in head and neck cancer and finds they outperform traditional radiomics and deep‑learning featu…