PyTorch: An Imperative Style, High-Performance Deep Learning Library

2019/12/03 by Adam Paszke, Paszke, Adam, Sam Gross +41 · 1 voice · 2303 citations
Computer Science · #Computational Physics and Python Applications #Machine Learning and Data Classification #Parallel Computing and Optimization Techniques #cs.LG #cs.MS #stat.ML

paper · pdf · doi:10.48550/arxiv.1912.01703

openalex publication_date 2019/12/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Deep learning frameworks have often focused on either usability or speed, but not both. PyTorch is a machine learning library that shows that these two goals are in fact compatible: it provides an imperative and Pythonic programming style that supports code as a model, makes debugging easy and is consistent with other popular scientific computing libraries, while remaining efficient and supporting hardware accelerators such as GPUs. In this paper, we detail the principles that drove the implementation of PyTorch and how they are reflected in its architecture. We emphasize that every aspect of PyTorch is a regular Python program under the full control of its user. We also explain how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance. We demonstrate the efficiency of individual subsystems, as well as the overall speed of PyTorch on several common benchmarks.

Cited by

• Evaluating Fuzz Testing for Reinforcement Learning Agents
• Learning to Rank for Selected Configuration Interaction
• Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
• AdamNX: An Adam improvement algorithm based on a novel exponential decay mechanism for the second-order moment estimate
• Forensics Adapter: Unleashing CLIP for Generalizable Face Forgery Detection
• Generalized Gaussian Temporal Difference Error for Uncertainty-aware Reinforcement Learning
• Universal Quantum Transformer
• SAGE-Net: Semantics-Augmented Geometric Encoder for Material Property Prediction
• An Introduction to Bayesian and Frequentist Simulation-Based Inference with Machine Learning
• Langevin for Nonconvex Optimization: Exact, Inexact and Zeroth-Order
• PolyGraphPy: A unified Python framework for atomistic simulation and machine learning-driven polymer design
• CARNet Cycle-Conditioned Core Aggregation and Redistribution for Multivariate Time Series Forecasting
• Flash EQ-Linear: Accelerating Equivariant Linear Layers via Group-wise Discrete Fourier Transform
• Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas
• A Weak Penalty Neural ODE for Learning Chaotic Dynamics from Noisy Time Series
• Hydrodynamic Backflow for Easing the Fermion Sign in Finite-Temperature Electron Path Integral Simulations
• Accounting for Hysteresis and Eddy Currents in Finite Element Simulations of Ferromagnetic Laminated Cores using a Recurrent Neural Network
• A Unified Variational Framework for Deep Weakly Supervised Image Segmentation
• Two-Step Occupation Coding
• Inferring activity from fluid flow in continuum models of active matter
• DGNA: Dissecting GPU NUMA Architecture through Microbenchmarking and Data Analysis
• Interactive Medical-SAM2 GUI: A Napari-based semi-automatic annotation tool for medical images
• Interval and fuzzy physics-augmented neural networks (iPANN and fPANN) for uncertainty quantification and propagation in constitutive modeling
• CoGoal3D: Collaborative 3D Object Detection with 3D-Aware Fusion and Refinement
• Adopting Reinforcement Learning with Verifiable Rewards for Molecular Generation
• Pre-Deployment Complexity Estimation for Federated Perception Systems
• DifFoundMAD: Foundation Models meet Differential Morphing Attack Detection
• Estimating the performance boundary of Gottesman-Kitaev-Preskill codes and number-phase codes
• LC-SLab -- An object-based deep learning framework for large-scale land cover classification from satellite imagery and sparse in-situ labels
• NVIDIA-labs OO Agents: Native Python Object-Oriented Agents
• End-to-End Learning of Safe Optimal Feedback Control in High Dimensions with Control Barrier Function Layers
• A real-time RGB-D perception pipeline for autonomous impact hammers in mining: self-filtering, rock segmentation and rock-breaking poses generation
• Machine-Learned Compact Subspace Generation for Quantum Selected Configuration Interaction within Density Matrix Embedding Framework
• Crowd4D: Scene-Aware Monocular 4D Crowd Reconstruction
• Reconstructing local environments from concise atomistic representations
• End-to-End Differential Privacy in Training Deep Neural Network Classifiers
• Beyond Post-Quantization: Native Hash Learning with a Dedicated HASH Token
• Posterior Samplings are Missing Modalities Generators for Medical Image Translation
• Cross-Modal UAV Object Tracking: State-Aware Representation Learning and A Unified Benchmark
• FlexiAvatar: Unified 3D Gaussian Human Avatars Under Arbitrary Body Visibility
• Towards anomaly detection searches for new physics signatures including Higgs bosons with weakly supervised machine learning
• IBoxCLA: Towards Robust Box-supervised Segmentation of Polyp via Improved Box-dice and Contrastive Latent-anchors
• Epistemic Familiarity is Associated With Belief Stability in Large Language Models
• SALT: Salience-Aware Lexical Trie for Long-Context Compression
• DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation
• LieBN: Batch Normalization over Lie Groups
• Selectivity Matters: Source Node Influence Pruning for Unsupervised Graph Domain Adaptation
• Benchmarking NACTI Species Recognition in Long-Tailed Regimes
• Fitting the topology of synthetic particle systems with a novel graph representation
• Program Synthesis for Simulation-Based Inference: Joint Model Selection and Parameter Estimation
• Efficient Tuning Before Low-Bit Post-Training Quantization for Stochastic Gradient Descent-optimized Models
• DirPA: Addressing Prior Shift in Imbalanced Few-shot Crop-type Classification
• Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning
• Certified Training for Convolutional Perturbations
• Budgeted Indirect Adversarial Attack on Graph-Based Anomaly Detection in Sensor Networks
• Tumor-anchored deep feature random forests for out-of-distribution detection in lung cancer segmentation
• Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak
• GQD-AdsNet: Graph Neural Networks Unlock Rapid Exploration of Transition Metal Adsorption on Graphene Quantum Dots
• Fine-Tuning a Universal Machine-Learned Interatomic Potential for Oxygen Plasma Interactions with WS₂
• Three-Body Scattering for Generative Modeling
• Retriever: Composing Closed-Loop Asynchronous Robot Programs
• Coordinated Disentanglement with Iterative Mode Discovery Under Hidden Correlations
• SubdivAR: Autoregressive Next-Scale Prediction for Neural Mesh Subdivision
• KEPLA: A Knowledge-Enhanced Deep Learning Framework for Accurate Protein-Ligand Binding Affinity Prediction
• Multi-scale closed-loop melt pool control for LPBF via policy optimization
• LATTICE: Graph Self-Supervised Learning for Multimodal Spatial Omics Integration
• Do We Really Need Quantum Machine Learning?: A Multidimensional Empirical Study
• Exact Network Surgery: Functional Invariance and Gradient Plasticity in Reactive Computational Graphs
• Evolution Strategy-Based Calibration for Low-Bit Quantization of Speech Models
• GNN-DIP: Neural Corridor Selection for Decomposition-Based Motion Planning
• It Takes a MAESTRO To Prune Bad Experts
• Cost Accounting for Reactive Computational Graphs: Exhaustive Sweeps, Sequential Mutation, and the Backward-Locality Gap
• Unsupervised Incremental Learning Using Confidence-Based Pseudo-Labels
• OpenLanguageModel: Readable and Composable Small-Language-Model Pretraining for Education and Research
• EEGPrep: a validated Python implementation of the EEGLAB preprocessing pipeline
• Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning
• Hybrid Machine Learning for Articulation Angle Estimation of Truck-Semitrailer Combinations
• Lightweight return-mapping surrogates for multiscale plasticity: a practical guide
• Spatio-Temporal Prediction of Unsteady Airfoil Aerodynamics Using Augmented Graph Neural Ordinary Differential Equations with Exogenous Controls
• NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction
• Every Microsecond Matters: Achieving Near Speed-of-Light Latency in GPU Collectives
• VAR-PZnn: A machine-learning framework for AGN photometric redshifts using color and variability-based features
• Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters
• Structure-Induced Information for Rerooting Levin Tree Search
• A Semiparametric Framework for Stochastic Fundamental Diagram Modeling
• A regularization method for quantum neural networks using data symmetry
• MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems
• Training a force field for proteins and small molecules from scratch
• LLM4EHR: Aligning Clinical Time Series with Medical Event Sequences via Large Language Models
• Differentiable Cardiac Electrophysiology Simulations for Dynamical State and Parameter Estimation
• Harnessing Machine Learning for Hybrid Constitutive Modelling of Viscoelastic Fluid Flows in Computational Rheology
• SMC-ES: Automated synthesis of formally verified control policies
• How large can galaxies be? Ultra-deep imaging of IC 1101, the most extended known galaxy
• Interleaved Noise Injection Improves Clean, Corrupted, and OOD Performance
• Unsupervised Evaluation of Deep Audio Embeddings for Music Structure Analysis
• GS-RealBlur: A Flexible Data Acquisition Framework for Real-World Image Deblurring
• Pretrained Event Classification Model for High Energy Physics Analysis
• Dendrite: A Real-Time Python Application for Online Brain-Computer Interface Research and Development
• Multiscale Mixed-Dimensional Simulation via Domain Decomposition and Non-Intrusive Neural Model Order Reduction
• VOiLA: Vectorized Online Planning with Learned Diffusion Models for POMDP Agents
• A Measurement Study of AI-Environment Realism Gaps in Malware-Analysis Sandboxes
• Missing Physics Discovery through Fully Differentiable Finite Element-Based Machine Learning
• MEDN: Motion-Emotion Feature Decoupling Network for Micro-Expression Recognition
• Molecular quantum control algorithm design by reinforcement learning
• Reliability-Aware Hard--Soft Physics-Informed Neural Networks for Robust Learning of Challenging Partial Differential Equations
• Representation recycling for streaming video analysis
• Manifold learning for source separation in confusion-limited gravitational-wave data
• TRACE: Trajectory-Based Safety Patch Learning for LLM Post-Training Realignment
• Token-Level Cross-Modal Transformer with Contrastive Multi-Task Learning for Breast Cancer Subtype Classification and Survival Prediction
• Proof-Carrying Multimodal Timelines: Finite-Trace Modal Certificates for Video-Audio Consistency
• Aircast-Mars: A Mars Foundation Model for Global Weather Forecasting with HEALPix-Aware Convolutions
• Jet quenching identification via supervised learning in simulated heavy-ion collisions
• MATNet: multi-level fusion transformer-based model for day-ahead PV generation forecasting
• Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
• Enhancing next token prediction based pre-training for jet foundation models
• The illusory simplicity of the feedforward pass: evidence for the dynamical nature of stimulus encoding along the primate ventral stream
• CURE: Privacy-Preserving Split Learning Done Right
• M²RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling
• All elementary functions from a single binary operator
• GrimACE: automated, multimodal cage-side assessment of pain and well-being in mice
• GPU-accelerated single-cell analysis at scale with rapids-singlecell
• Deep models of protein evolution in time generate realistic evolutionary trajectories and functional proteins
• Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network
• 2026 International Conference on 3D Vision (3DV)
• Benchmarking for practice: Few-shot time-series crop-type classification on the EuroCropsML dataset
• Multi-Agent Inverted Transformer for Flight Trajectory Prediction
• Bundle Adjustment in the Eager Mode
• On the use of chemical bonding descriptors in machine learning
• Predicting upcoming visual features during eye movements yields scene representations aligned with human visual cortex
• Taming diffusion transformers for high-fidelity MRI super-resolution
• Modelling transcription with explainable AI uncovers context-specific epigenetic gene regulation at promoters and gene bodies
• Expected Attention: KV Cache Compression by Estimating Attention from Future Queries Distribution
• Global, multi-scale standing deadwood segmentation in centimeter-scale aerial images
• Generalized Orders of Magnitude for Scalable, Parallel, High-Dynamic-Range Computation
• scPortrait integrates single-cell images into multimodal modeling
• Flexible inference for animal learning rules using neural networks
• Flashzoi: an enhanced Borzoi for accelerated genomic analysis
• Biophysics-based protein language models for protein engineering
• A realistic phantom dataset for benchmarking cryo-ET data annotation
• SplitQuantV2: Enhancing Low-Bit Quantization of LLMs Without GPUs
• Cross‐Dataset Evaluation of Dementia Longitudinal Progression Prediction Models
• Partial recurrence enables robust and efficient computation
• A Sitewise Model of Natural Selection on Individual Antibodies via a Transformer–Encoder
• Supervised Fine Tuning on Curated Data is Reinforcement Learning (and can be improved)
• Real-Time, Inline Quantitative MRI Enabled by Scanner-Integrated Machine Learning: A Proof of Principle With NODDI.
• Relative Entropy Pathwise Policy Optimization
• Deep Sigma Point Processes for RCS Modeling in Spaceborne SAR Imagery
• ShapeEmbed: a self-supervised learning framework for 2D contour quantification
• Hardware Accelerated Neural Block Texture Compression with Cooperative Vectors
• Self-supervised learning of molecular representations from millions of tandem mass spectra using DreaMS
• Quick ViTs: Speeding up Vision Transformers through Equivariance
• CaRL: Learning Scalable Planning Policies with Simple Rewards
• Learning Dynamics of RNNs in Closed-Loop Environments
• Toward a Sparse and Interpretable Audio Codec
• A Common Interface for Automatic Differentiation
• Scalable and Performant Data Loading
• Pushing the Accuracy Limit of Foundation Neural Network Models with Quantum Monte Carlo Forces and Path Integrals
• AssistanceZero: Scalably Solving Assistance Games
• NeuRaLaTeX: A machine learning library written in pure LaTeX
• Inverse problems with experiment-guided AlphaFold
• Reevaluating Policy Gradient Methods for Imperfect-Information Games
• Extending the RANGE of Graph Neural Networks: Relaying Attention Nodes for Global Encoding
• Smaller But Better: Unifying Layout Generation with Smaller Large Language Models
• Slow cortical dynamics generate context processing and novelty detection
• ILIAS: Instance-Level Image retrieval At Scale
• Cover Learning for Large-Scale Topology Representation
• The Cake that is Intelligence and Who Gets to Bake it: An AI Analogy and its Implications for Participation
• The functional role of oscillatory dynamics in neocortical circuits: A computational perspective
• DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
• Sparser, Better, Faster, Stronger: Sparsity Detection for Efficient Automatic Differentiation
• Bi-Level Decision-Focused Causal Learning for Large-Scale Marketing Optimization: Bridging Observational and Experimental Data
• On the accuracy of implicit neural representations for cardiovascular anatomies and hemodynamic fields
• TorchQuantumDistributed
• LoMa: Local Feature Matching Revisited
• Measuring complex constructs in large-scale text with computational social mixed methods
• FAMUS: A Few-Shot Learning Framework for Large-Scale Protein Annotation
• NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models
• Deterministic Image-to-Image Translation via Denoising Brownian Bridge Models with Dual Approximators
• FcaNet: Frequency Channel Attention Networks
• AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis
• Graph Set Transformer
• Rethinking Intrinsic Dimension Estimation in Neural Representations
• A Probabilistic Framework for LLM-Based Model Discovery
• Stochastic tensor contraction for quantum chemistry
• A new adaptive two-layer model for opinion spread in hypergraphs: parameter sensitivity and estimation
• Reservoir Computing inspired Matrix Multiplication-free Language Model
• Is Chain-of-Thought Really Not Explainability? Chain-of-Thought Can Be Faithful without Hint Verbalization
• Credit-assigned Policy Gradient for Early Stage Retrieval in Two-stage Ranking
• Similar destabilization of neural dynamics under different general anesthetics
• Galaxy Zoo Evo: 1 million human-annotated images of galaxies
• Let Samples Speak: Mitigating Spurious Correlation by Exploiting the Clusterness of Samples
• What Matters in Deep Learning for Time Series Forecasting?
• Modality Inflation: Energy Characterization and Optimization Opportunities for MLLM Inference
• Tree Meets Transformer: A Hybrid Architecture for Scalable Power Allocation in Cell-Free Networks
• TimePerceiver: An Encoder-Decoder Framework for Generalized Time-Series Forecasting
• LuxIA: A Lightweight Unitary matriX-based Framework Built on an Iterative Algorithm for Photonic Neural Network Training
• SimBEV2X: A Large-Scale Dataset and Data Generation Tool for Multi-Task Vehicle-to-Everything Cooperative Perception
• Agent Data Injection Attacks are Realistic Threats to AI Agents
• Gaze-to-text Generation: Beyond Categorical Decoding of Human Attention
• DDVT: Dynamic Dual-level Vision Transformer Fusion Network for Answer Grounding in Visual Question Answering
• Minimum Distance Summaries for Robust Neural Posterior Estimation
• Extending Fourier Neural Operators for Modeling Parameterized and Coupled PDEs
• Weakly Supervised Instance-Level Gleason Pattern Estimation Using Primary and Secondary Labels
• Robust 6-DoF Object Pose Tracking with Built-In Recovery under Occlusions and Rapid Object Motions
• Full event interpretation with machine-learning-based particle-flow reconstruction in the CMS detector
• Identifying Dolphin Whistle Producers With Deep Learning: Moving Beyond Signature Whistles
• Bitcoin Price Direction Prediction via Regime-Aware Multi-Modal Fusion of Social Sentiment and Technical Features
• Scale Weight Decay and Train Better
• Navigating protein landscapes with a machine-learned transferable coarse-grained model
• Deep material networks for fiber suspensions with infinite material contrast
• MEGA-CL: A Molecular Foundation Model for Generalizable ADMET Prediction through Graph External Attention and Contrastive Learning
• UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models
• SPRKD: Effective Knowledge Distillation for Deep Neural Networks via Saddle Region Approximation
• EXE-Bench: Ranking the Tradeoffs of AI-based Windows Malware Detectors for Real-World Usability
• Looking for Affect in Spontaneous Finnish Speech through Linguistic Interpretability
• Calibrated Tree-Neural Fusion for Fine-Grained Vegetation Community Classification
• DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning
• Optimized Embedded Implementation of Hyperspectral-Multispectral Image Fusion on Raspberry Pi
• Breaking the Synthetic-Real Domain Shortcut for Training-Free Generative Replay-based Class Incremental Learning
• A Reference-Free Framework for Evaluating Single-Frame ISP Pipelines
• Automated Numerical Stability Analysis of Deep Learning Operators
• Retraction-Free Optimization over the Stiefel Manifold for the LoRA Fine-Tuning
• DensFiLM: Density-Conditioned Video Saliency for Crowd Scenes
• Leak-Free Cross-Validated Stacking with Per-Architecture Calibration for Sand-Boil Segmentation in Earthen Levees
• Fast, accurate, and differentiable: a neural-network surrogate for NRSur7dq4 precessing binary black hole waveforms
• Multimodal Surface EMG Hand Gesture Recognition Using Query-Based Transformers for Prosthetic Control
• Controlling Embedding Spaces with Text-Conditioned Transformations
• Real-time Reconstruction of Human Visual Perception from fMRI
• Statistical Mechanics of Thermal Diffusion in Rough Energy Landscapes
• Benchmarking the Domain Gap: Model Selection Instability Under Domain Shift in Video Capsule Endoscopy
• Symmetry Alone Is Not an Ansatz: Task-Aligned Interactions in Equivariant Quantum Circuits
• Detect Before You Leap: Mirage Detection in Vision-Language Models
• polyDAG: Polynomial Acyclicity Constraints for Efficient Continuous Causal Discovery in Visual Semantic Graphs
• DynaResize: Runtime GPU Reallocation for Disaggregated LLM Post-Training
• Quotient Tree Arithmetic: Deferred-Division Computation with Bounded Symbolic Depth and Cross-Subtree Cancellation
• xMIx: High-Performance Serving-Time Platform for Mechanistic Interpretability Apps
• Extracting Exact Lie Derivatives Without Backpropagation: A Dual Compiler for Neural Control Barrier Functions
• Kernel Forge: An Agent Harness for LLM-based Generation and Optimization of CUDA Kernels
• Bridging MARL to SARL: An Order-Independent Multi-Agent Transformer via Latent Consensus
• Unraveling the Mechanism of Drug Binding to SARS‐CoV‐2 RNA Pseudoknot With Thermodynamics‐Driven Machine Learning
• Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex
• Domain Decomposition of Large Neural Network Surrogate Models
• Plain Transformers are Surprisingly Powerful Link Predictors
• Calibrating adaptive smoothing methods for freeway traffic reconstruction
• FUSCO: High-Performance Distributed Data Shuffling via Transformation-Communication Fusion
• Batched Training for QLSTM vs. QFWP: A System-Oriented Approach to EPC-Aware RMSE-DA
• Compliance Rating Scheme: A Data Provenance Framework for Generative AI Datasets
• Contrastive Graph Modeling for Cross-Domain Few-Shot Medical Image Segmentation
• UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion Segmentation
• AVP-Fusion: Adaptive Multi-Modal Fusion and Contrastive Learning for Two-Stage Antiviral Peptide Identification
• Global-Graph Guided and Local-Graph Weighted Contrastive Learning for Unified Clustering on Incomplete and Noise Multi-View Data
• kooplearn: A Scikit-Learn Compatible Library of Algorithms for Evolution Operator Learning
• Parallel Token Prediction for Language Models
• Variationally correct operator learning: Reduced basis neural operator with a posteriori error estimation
• TGC-Net: A Structure-Aware and Semantically-Aligned Framework for Text-Guided Medical Image Segmentation
• SparScene: Efficient Traffic Scene Representation via Sparse Graph Learning for Large-Scale Trajectory Generation
• Blurb-Refined Inference from Crowdsourced Book Reviews using Hierarchical Genre Mining with Dual-Path Graph Convolutions
• Language-Guided Grasp Detection with Coarse-to-Fine Learning for Robotic Manipulation
• Linear Attention for Joint Power Optimization and User-Centric Clustering in Cell-Free Networks
• Diving into 3D Parallelism with Heterogeneous Spot Instance GPUs: Design and Implications
• MMSRARec: Summarization and Retrieval Augumented Sequential Recommendation Based on Multimodal Large Language Model
• Embodied AI-Enhanced IoMT Edge Computing: UAV Trajectory Optimization and Task Offloading with Mobility Prediction
• Equivariant Multiscale Learned Invertible Reconstruction for Cone Beam CT: From Simulated to Real Data
• RHAPSODY: Execution of Hybrid AI-HPC Workflows at Scale
• ASCHOPLEX encounters Dafne: a federated continuous learning project for the generalizability of the Choroid Plexus automatic segmentation
• Towards Generative Location Awareness for Disaster Response: A Probabilistic Cross-view Geolocalization Approach
• Programmable Optical Spectrum Shapers as Computing Primitives for Accelerating Convolutional Neural Networks
• Neural Scaling Laws for Learning-based Identification of Nonlinear Systems
• A Comprehensive Study of Bugs in Modern Distributed Deep Learning Systems
• Debate-Enhanced Pseudo Labeling and Frequency-Aware Progressive Debiasing for Weakly-Supervised Camouflaged Object Detection with Scribble Annotations
• Optimality-Informed Neural Networks for Solving Parametric Optimization Problems
• JDPNet: A Network Based on Joint Degradation Processing for Underwater Image Enhancement
• SHIRO: Near-Optimal Communication Strategies for Distributed Sparse Matrix Multiplication
• H²em: Learning Hierarchical Hyperbolic Embeddings for Compositional Zero-Shot Learning
• Gaussian Process Assisted Meta-learning for Image Classification and Object Detection Models
• Neuron-Guided Interpretation of Code LLMs: Where, Why, and How?
• Spectral Diffusion for Sampling on \rm SU(N)
• Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function
• Kolmogorov-Arnold Graph Neural Networks Applied to Inorganic Nanomaterials Dataset
• LeLaR: The First In-Orbit Demonstration of an AI-Based Satellite Attitude Controller
• Dynamic Stream Network for Combinatorial Explosion Problem in Deformable Medical Image Registration
• GLUE: Coordinating Pre-Trained Generative Models for System-Level Design
• Learning General Policies with Policy Gradient Methods
• Evidential Trust-Aware Model Personalization in Decentralized Federated Learning for Wearable IoT
• Specific Multi-emitter Identification: Theoretical Limits and Low-complexity Design
• Dual Model Deep Learning for Alzheimer Prognostication
• Steering Vision-Language Pre-trained Models for Incremental Face Presentation Attack Detection
• DeepQuantum: A PyTorch-based Software Platform for Quantum Machine Learning and Photonic Quantum Computing
• Self-Attention with State-Object Weighted Combination for Compositional Zero Shot Learning
• Adaptive Probability Flow Residual Minimization for High-Dimensional Fokker-Planck Equations
• Non-stationary Spatial Modeling Using Fractional SPDEs
• Text2Graph VPR: A Text-to-Graph Expert System for Explainable Place Recognition in Changing Environments
• Multi-Part Object Representations via Graph Structures and Co-Part Discovery
• When Does Learning Renormalize? Sufficient Conditions for Power Law Spectral Dynamics
• FedWiLoc: Federated Learning for Privacy-Preserving WiFi Indoor Localization
• ALIGN: Advanced Query Initialization with LiDAR-Image Guidance for Occlusion-Robust 3D Object Detection
• Transfer Learning for Analysis of Collective and Non-Collective Thomson Scattering Spectra
• Optimal Software Pipelining and Warp Specialization for Tensor Core GPUs
• InfinityEBSD : Metrics-Guided Infinite-Size EBSD Map Generation With Diffusion Models
• Implicit Likelihood Inference of the Neutrino Mass Hierarchy from Cosmological Data
• GreedySnake: Accelerating SSD-Offloaded LLM Training with Efficient Scheduling and Optimizer Step Overlapping
• Validation of Diagnostic Artificial Intelligence Models for Prostate Pathology in a Middle Eastern Cohort
• Generative modeling of conditional probability distributions on the level-sets of collective variables
• Beyond Semantic Features: Pixel-level Mapping for Generalized AI-Generated Image Detection
• Guided progressive reconstructive imaging: a new quantization-based framework for low-dose, high-throughput and real-time analytical ptychography
• MIRGE: An Array-Based Computational Framework for Scientific Computing
• XLM: A Python package for non-autoregressive language models
• Adversarial VR: An Open-Source Testbed for Evaluating Adversarial Robustness of VR Cybersickness Detection and Mitigation
• SFTok: Bridging the Performance Gap in Discrete Tokenizers
• In-Context Algebra
• A Cartesian-3j Framework for Machine Learning Interatomic Potentials
• Semi-Supervised Online Learning on the Edge by Transforming Knowledge from Teacher Models
• PrivateXR: Defending Privacy Attacks in Extended Reality Through Explainable AI-Guided Differential Privacy
• NRGPT: An Energy-based Alternative for GPT
• DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI
• Efficient CPU-GPU Collaborative Inference for MoE-based LLMs on Memory-Limited Systems
• AI4EOSC: a Federated Cloud Platform for Artificial Intelligence in Scientific Research
• LaverNet: Lightweight All-in-one Video Restoration via Selective Propagation
• DAG Learning from Zero-Inflated Count Data Using Continuous Optimization
• Dual-View Inference Attack: Machine Unlearning Amplifies Privacy Exposure
• DRIVE: One-bit Distributed Mean Estimation
• Metanetworks as Regulatory Operators: Learning to Edit for Requirement Compliance
• Machine Learning Enabled Graph Analysis of Particulate Composites: Application to Solid-state Battery Cathodes
• dLITE: Differentiable Lighting-Informed Trajectory Evaluation for On-Orbit Inspection
• Higher-Order LaSDI: Reduced Order Modeling with Multiple Time Derivatives
• SALVE: Sparse Autoencoder-Latent Vector Editing for Mechanistic Control of Neural Networks
• Physics-informed Neural Operators for Predicting 3D Electromagnetic Fields Transformed by Metasurfaces
• Low-Latency FPGA Control System for Real-Time Neural Network Processing in CCD-Based Trapped-Ion Qubit Measurement
• Enhancing Alzheimer's Detection through Late Fusion of Multi-Modal EEG Features
• An Efficient Gradient-Based Inference Attack for Federated Learning
• Neural Modular Physics for Elastic Simulation
• Meta-learners for few-shot weakly-supervised optic disc and cup segmentation on fundus images
• NAP3D: NeRF Assisted 3D-3D Pose Alignment for Autonomous Vehicles
• Self-adaptive physics-informed neural network for forward and inverse problems in heterogeneous porous flow
• Composition-agnostic prediction of self-assembly in multicomponent amphiphile mixtures from molecular structure
• Focus: A Streaming Concentration Architecture for Efficient Vision-Language Models
• Hybrid Iterative Solvers with Geometry-Aware Neural Preconditioners for Parametric PDEs
• Search for heavy neutral leptons in B-meson decays
• FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting
• Moment-Based 3D Gaussian Splatting: Resolving Volumetric Occlusion with Order-Independent Transmittance
• Understanding and Improving Hyperbolic Deep Reinforcement Learning
• FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation
• Learning Minimal Representations of Fermionic Ground States
• Towards Test-time Efficient Visual Place Recognition via Asymmetric Query Processing
• TorchTraceAP: A New Benchmark Dataset for Detecting Performance Anti-Patterns in Computer Vision Models
• Consistent Instance Field for Dynamic Scene Understanding
• ProtoFlow: Interpretable and Robust Surgical Workflow Modeling with Learned Dynamic Scene Graph Prototypes
• SonicMoE: Accelerating MoE with IO and Tile-aware Optimizations
• Machine learning discovers new champion codes
• Nexels: Neurally-Textured Surfels for Real-Time Novel View Synthesis with Sparse Geometries
• Directional Textual Inversion for Personalized Text-to-Image Generation
• Fast Policy Learning for 6-DOF Position Control of Underwater Vehicles
• Face Identity Unlearning for Retrieval via Embedding Dispersion
• Measurement of Material Volume Fractions in a Microwave Resonant Cavity Sensor Using Convolutional Neural Network
• StarryGazer: Leveraging Monocular Depth Estimation Models for Domain-Agnostic Single Depth Image Completion
• Alada: Alternating Adaptation of Momentum Method for Memory-Efficient Matrix Optimization
• FlashFuser: Expanding the Scale of Kernel Fusion for Compute-Intensive Operators via Inter-Core Connection
• Towards High-Fidelity Gaussian Splatting with Queried-Convolution Neural Networks
• Practical Hybrid Quantum Language Models with Observable Readout on Real Hardware
• Anatomy-Guided Representation Learning Using a Transformer-Based Network for Thyroid Nodule Segmentation in Ultrasound Images
• GrowTAS: Progressive Expansion from Small to Large Subnets for Efficient ViT Architecture Search
• Exploring Spatial-Temporal Representation via Star Graph for mmWave Radar-based Human Activity Recognition
• AI Benchmark Democratization and Carpentry
• FRQI Pairs method for image classification using Quantum Recurrent Neural Network
• TopicProphet: Prophesies on Temporal Topic Trends and Stocks
• Type II and Type III Solar Radio Burst Classification Using Transfer Learning
• NoveltyRank: A Retrieval-Augmented Framework for Conceptual Novelty Estimation in AI Research
• FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model
• Machine learned potential for defected single layer hexagonal boron nitride
• Multi-task Learning with Extended Temporal Shift Module for Temporal Action Localization
• SUMFORU: An LLM-Based Review Summarization Framework for Personalized Purchase Decision Support
• In-Context Multi-Objective Optimization
• Data-Driven Model Reduction using WeldNet: Windowed Encoders for Learning Dynamics
• Interpretable and Steerable Concept Bottleneck Sparse Autoencoders
• Video Depth Propagation
• Topology-Guided Quantum GANs for Constrained Graph Generation
• Robust Shape from Focus via Multiscale Directional Dilated Laplacian and Recurrent Network
• Semantic Reconstruction of Adversarial Plagiarism: A Context-Aware Framework for Detecting and Restoring "Tortured Phrases" in Scientific Literature
• D2M: A Decentralized, Privacy-Preserving, Incentive-Compatible Data Marketplace for Collaborative Learning
• Point2Pose: A Generative Framework for 3D Human Pose Estimation with Multi-View Point Cloud Dataset
• High-Dimensional Data Processing: Benchmarking Machine Learning and Deep Learning Architectures in Local and Distributed Environments
• Design Space Exploration of DMA based Finer-Grain Compute Communication Overlap
• Federated Domain Generalization with Latent Space Inversion
• THeGAU: Type-Aware Heterogeneous Graph Autoencoder and Augmentation
• Uncertainty-Preserving QBNNs: Multi-Level Quantization of SVI-Based Bayesian Neural Networks for Image Classification
• Straggler Tolerant and Resilient DL Training on Homogeneous GPUs
• Neuromorphic Eye Tracking for Low-Latency Pupil Detection
• Transport Novelty Distance: A Distributional Metric for Evaluating Material Generative Models
• RACAM: Enhancing DRAM with Reuse-Aware Computation and Automated Mapping for ML Inference
• A Hybrid Neural Network-Finite Element Method for the Viscous-Plastic Sea-Ice Model
• Prompt-Based Continual Compositional Zero-Shot Learning
• SDialog: A Python Toolkit for End-to-End Agent Building, User Simulation, Dialog Generation, and Evaluation
• Language-Conditioned Safe Trajectory Generation for Spacecraft Rendezvous
• Modular Deep-Learning-Based Early Warning System for Deadly Heatwave Prediction
• Quantum Decision Transformers (QDT): Synergistic Entanglement and Interference for Offline Reinforcement Learning
• Attention is All You Need to Defend Against Indirect Prompt Injection Attacks in LLMs
• Magneton: Optimizing Energy Efficiency of ML Systems via Differential Energy Debugging
• Age-Inclusive 3D Human Mesh Recovery for Action-Preserving Data Anonymization
• When unlearning is free: leveraging low influence points to reduce computational costs
• Closed-Loop Robotic Manipulation of Transparent Substrates for Self-Driving Laboratories using Deep Learning Micro-Error Correction
• Probabilistic Multi-Agent Aircraft Landing Time Prediction
• A Bayesian latent class reinforcement learning framework to capture adaptive, feedback-driven travel behaviour
• SSplain: Sparse and Smooth Explainer for Retinopathy of Prematurity Classification
• Forecasting Dark Matter Subhalo Constraints from Stellar Streams using Implicit Likelihood Inference
• Formalized Hopfield Networks and Boltzmann Machines
• sim2art: Accurate Articulated Object Modeling from a Single Video using Synthetic Training Data Only
• Delay-Aware Diffusion Policy: Bridging the Observation-Execution Gap in Dynamic Tasks
• A Geometric Unification of Concept Learning with Concept Cones
• Generalized Referring Expression Segmentation on Aerial Photos
• Comparing BFGS and OGR for Second-Order Optimization
• Object Pose Distribution Estimation for Determining Revolution and Reflection Uncertainty in Point Clouds
• Materium: An Autoregressive Approach for Material Generation
• Machine Learning-based Unfolding for Cross Section Measurements in the Presence of Nuisance Parameters
• DAUNet: A Lightweight UNet Variant with Deformable Convolutions and Parameter-Free Attention for Medical Image Segmentation
• Investigating Training and Generalization in Faithful Self-Explanations of Large Language Models
• SparsePixels: Efficient Convolution for Sparse Data on FPGAs
• Large Language Model-Based Generation of Discharge Summaries
• TreeQ: Pushing the Quantization Boundary of Diffusion Transformer via Tree-Structured Mixed-Precision Search
• Modeling Spatio-temporal Extremes via Conditional Variational Autoencoders
• FEALPy v3: A Cross-platform Intelligent Numerical Simulation Engine
• Learning Relative Gene Expression Trends from Pathology Images in Spatial Transcriptomics
• EMGauss: Continuous Slice-to-3D Reconstruction via Dynamic Gaussian Modeling in Volume Electron Microscopy
• A Perception CNN for Facial Expression Recognition
• AgenticCyber: A GenAI-Powered Multi-Agent System for Multimodal Threat Detection and Adaptive Response in Cybersecurity
• JaxWildfire: A GPU-Accelerated Wildfire Simulator for Reinforcement Learning
• Empowering GNNs for Domain Adaptation via Denoising Target Graph
• GPU-GLMB: Assessing the Scalability of GPU-Accelerated Multi-Hypothesis Tracking
• From Remote Sensing to Multiple Time Horizons Forecasts: Transformers Model for CyanoHAB Intensity in Lake Champlain
• ShadowWolf -- Automatic Labelling, Evaluation and Model Training Optimised for Camera Trap Wildlife Images
• A Latent Variable Framework for Scaling Laws in Large Language Models
• Physics-Grounded Attached Shadow Detection Using Approximate 3D Geometry and Light Direction
• Embedding Physical Reasoning into Diffusion-Based Shadow Generation
• Automated Annotation of Shearographic Measurements Enabling Weakly Supervised Defect Detection
• Shoot-Bounce-3D: Single-Shot Occlusion-Aware 3D from Lidar by Decomposing Two-Bounce Light
• SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models
• KQ-SVD: Compressing the KV Cache with Provable Guarantees on Attention Fidelity
• EventQueues: Autodifferentiable spike event queues for brain simulation on AI accelerators
• DAE-HardNet: A Physics Constrained Neural Network Enforcing Differential-Algebraic Hard Constraints
• Consist-Retinex: One-Step Noise-Emphasized Consistency Training Accelerates High-Quality Retinex Enhancement
• Big Tech-Funded AI Papers Have Higher Citation Impact, Greater Insularity, and Larger Recency Bias
• Bounded Graph Clustering with Graph Neural Networks
• Wasserstein distance based semi-supervised manifold learning and application to GNSS multi-path detection
• Hybrid modeling approach for better identification of building thermal network model and improved prediction
• Rep Smarter, Not Harder: AI Hypertrophy Coaching with Wearable Sensors and Edge Neural Networks
• Meta-Learning for Quantum Optimization via Quantum Sequence Model
• QKAN-LSTM: Quantum-inspired Kolmogorov-Arnold Long Short-term Memory
• An Efficient and Accurate Surrogate Modeling of Flapping Dynamics in Inverted Elastic Foils using Hypergraph Neural Networks
• Efficient Generative Transformer Operators For Million-Point PDEs
• STELLA: Guiding Large Language Models for Time Series Forecasting with Semantic Abstractions
• SoftJAX & SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients
• Multiple Source Localization via Local Radio Map Construction in Urban Environments
• Recurrent Neural Networks with Linear Structures for Electricity Price Forecasting
• A Footprint-Aware, High-Resolution Approach for Carbon Flux Prediction Across Diverse Ecosystems
• UTrice: Unifying Primitives in Differentiable Ray Tracing and Rasterization via Triangles for Particle-Based 3D Scenes
• ReflexFlow: Rethinking Learning Objective for Exposure Bias Alleviation in Flow Matching
• OnSight Pathology: A real-time platform-agnostic computational pathology companion for histopathology
• Digital Twin-based Control Co-Design of Full Vehicle Active Suspensions via Deep Reinforcement Learning
• Hybrid Temporal-8-Bit Spike Coding for Spiking Neural Network Surrogate Training
• Studying Various Activation Functions and Non-IID Data for Machine Learning Model Robustness
• PosA-VLA: Enhancing Action Generation via Pose-Conditioned Anchor Attention
• GaussianBlender: Instant Stylization of 3D Gaussians with Disentangled Latent Spaces
• Multi-Scale Visual Prompting for Lightweight Small-Image Classification
• FFTrainer: Fast Failover in Large-Language Model Training with Almost-Free State Management
• Harnessing Hypergraphs in Geometric Deep Learning for 3D RNA Inverse Folding
• Resilient AFE Drive Control using Neural Networks with Tracking Guarantees
• Agentic Operator Generation for ML ASICs
• VS-Graph: Scalable and Efficient Graph Classification Using Hyperdimensional Computing
• FuXi-Nowcast: Environment-conditioned deep learning for severe convection nowcasting
• Flux4D: Flow-based Unsupervised 4D Reconstruction
• Learning interpretable surface elasticity properties from bulk properties
• U4D: Uncertainty-Aware 4D World Modeling from LiDAR Sequences
• Flexible Gravitational-Wave Parameter Estimation with Transformers
• PAC-Bayesian Optimal Control with Stability and Generalization Guarantees
• PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models
• Cross-Lingual Prompt Steerability: Towards Accurate and Robust LLM Behavior across Languages
• Dynamic Configuration of On-Street Parking Spaces using Multi Agent Reinforcement Learning
• Study of fully coupled 3D envelope instability using automatic differentiation
• Enhancing Floor Plan Recognition: A Hybrid Mix-Transformer and U-Net Approach for Precise Wall Segmentation
• Sampling on Metric Graphs
• Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design
• Unifying Sign and Magnitude for Optimizing Deep Vision Networks via ThermoLion
• BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch
• On the Unreasonable Effectiveness of Last-layer Retraining
• Comparing Baseline and Day-1 Diffusion MRI Using Multimodal Deep Embeddings for Stroke Outcome Prediction
• Morphling: Fast, Fused, and Flexible GNN Training at Scale
• SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge
• Accelerated Machine Learning Force Field for Predicting Thermal Conductivity of Organic Liquids
• Neural Networks for Predicting Permeability Tensors of 2D Porous Media: Comparison of Convolution- and Transformer-based Architectures
• zea: A Toolbox for Cognitive Ultrasound Imaging
• Textured Geometry Evaluation: Perceptual 3D Textured Shape Metric via 3D Latent-Geometry Network
• Learning to Reconstruct Temperature Field from Sparse Observations with Implicit Physics Priors
• mmPred: Radar-based Human Motion Prediction in the Dark
• Learning Eigenstructures of Unstructured Data Manifolds
• Upper Approximation Bounds for Neural Oscillators
• Feed-Forward 3D Gaussian Splatting Compression with Long-Context Modeling
• Smol-GS: Compact Representations for Abstract 3D Gaussian Splatting
• CAR-Net: A Cascade Refinement Network for Rotational Motion Deblurring under Angle Information Uncertainty
• Privacy Preserving Diffusion Models for Mixed-Type Tabular Data Generation
• DialBench: Towards Accurate Reading Recognition of Pointer Meter using Large Foundation Models
• PAT3D: Physics-Augmented Text-to-3D Scene Generation
• Faster Verified Explanations for Neural Networks
• FPGA-Accelerated Real-Time Beam Emission Spectroscopy Diagnostics at DIII-D Using the SLAC Neural Network Library for ML Inference
• Search for H→ cc and measurement of H→ bb in vector-boson fusion production with the ATLAS Detector
• DEAL-300K: Diffusion-based Editing Area Localization with a 300K-Scale Dataset and Frequency-Prompted Baseline
• PointCNN++: Performant Convolution on Native Points
• RobotSeg: A Model and Dataset for Segmenting Robots in Image and Video
• DenoiseGS: Gaussian Reconstruction Model for Burst Denoising
• Switching-time bioprocess control with pulse-width-modulated optogenetics
• db-SP: Accelerating Sparse Attention for Visual Generative Models with Dual-Balanced Sequence Parallelism
• TARFVAE: Efficient One-Step Generative Time Series Forecasting via TARFLOW based VAE
• Manifolds and Modules: How Function Develops in a Neural Foundation Model
• Saddle-Free Guidance: Improved On-Manifold Sampling without Labels or Additional Training
• A multi-language auto-differentiation module and its application to a parallel particle-in-cell code on distributed computers
• Adversarial Flow Models
• ITS3D: Inference-Time Scaling for Text-Guided 3D Diffusion Models
• Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
• SingleQuant: Efficient Quantization of Large Language Models in a Single Pass
• Shoe Style-Invariant and Ground-Aware Learning for Dense Foot Contact Estimation
• Softly Symbolifying Kolmogorov-Arnold Networks
• Waveform-Based Probabilistic Seismic Hazard Analysis Using Ground-Motion Generative Models
• Self-Paced Learning for Images of Antinuclear Antibodies
• SAM Guided Semantic and Motion Changed Region Mining for Remote Sensing Change Captioning
• Hierarchical Ranking Neural Network for Long Document Readability Assessment
• Mean-Field Limits for Two-Layer Neural Networks Trained with Consensus-Based Optimization
• EvRainDrop: HyperGraph-guided Completion for Effective Frame and Event Stream Aggregation
• Subjective Depth and Timescale Transformers: Learning Where and When to Compute
• GPU Memory Prediction for Multimodal Model Training
• RefOnce: Distilling References into a Prototype Memory for Referring Camouflaged Object Detection
• Open Vocabulary Compositional Explanations for Neuron Alignment
• Active learning with physics-informed neural networks for optimal sensor placement in deep tunneling through transversely isotropic elastic rocks
• Gated Uncertainty-Aware Runtime Dual Invariants for Neural Signal-Controlled Robotics
• 3D-Aware Multi-Task Learning with Cross-View Correlations for Dense Scene Understanding
• Automated Monitoring of Cultural Heritage Artifacts Using Semantic Segmentation
• Identifying environmental factors associated with tetrodotoxin contamination in bivalve mollusks using eXplainable AI
• MXtalTools: A Toolkit for Machine Learning on Molecular Crystals
• Beyond Components: Singular Vector-Based Interpretability of Transformer Circuits
• Advancing Image Classification with Discrete Diffusion Classification Modeling
• XiCAD: Camera Activation Detection in the Da Vinci Xi User Interface
• QiMeng-Kernel: Macro-Thinking Micro-Coding Paradigm for LLM-Based High-Performance GPU Kernel Generation
• Low-Resolution Editing is All You Need for High-Resolution Editing
• Clair Obscur: an Illumination-Aware Method for Real-World Image Vectorization
• Efficient Optimization of a Permanent Magnet Array for a Stable 2D Trap
• Accuracy and Efficiency Trade-Offs in LLM-Based Malware Detection and Explanation: A Comparative Study of Parameter Tuning vs. Full Fine-Tuning
• BackSplit: The Importance of Sub-dividing the Background in Biomedical Lesion Segmentation
• DensifyBeforehand: LiDAR-assisted Content-aware Densification for Efficient and Quality 3D Gaussian Splatting
• DynaMix: Generalizable Person Re-identification via Dynamic Relabeling and Mixed Data Sampling
• ModHiFi: Identifying High Fidelity predictive components for Model Modification
• Resolving Node Identifiability in Graph Neural Processes via Laplacian Spectral Encodings
• TPG-INR: Target Prior-Guided Implicit 3D CT Reconstruction for Enhanced Sparse-view Imaging
• Higgs Production Classifier using Weak Supervision
• DriveFlow: Rectified Flow Adaptation for Robust 3D Object Detection in Autonomous Driving
• Modality-Collaborative Low-Rank Decomposers for Few-Shot Video Domain Adaptation
• Neural Geometry Image-Based Representations with Optimal Transport (OT)
• Subtract the Corruption: Training-Data-Free Corrective Machine Unlearning using Task Arithmetic
• Peregrine: One-Shot Fine-Tuning for FHE Inference of General Deep CNNs
• VecIntrinBench: Benchmarking Cross-Architecture Intrinsic Code Migration for RISC-V Vector
• A Systematic Study of Compression Ordering for Large Language Models
• A Novel and Practical Universal Adversarial Perturbations against Deep Reinforcement Learning based Intrusion Detection Systems
• stable-pretraining-v1: Foundation Model Research Made Simple
• Developing an AI Course for Synthetic Chemistry Students
• Hyperspectral Variational Autoencoders for Joint Data Compression and Component Extraction
• Crash-Consistent Checkpointing for AI Training on macOS/APFS
• Transforming Conditional Density Estimation Into a Single Nonparametric Regression Task
• MimiCAT: Mimic with Correspondence-Aware Cascade-Transformer for Category-Free 3D Pose Transfer
• scipy.spatial.transform: Differentiable Framework-Agnostic 3D Transformations in Python
• Hybrid Event Frame Sensors: Modeling, Calibration, and Simulation
• GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models
• Learning Rate Scheduling with Matrix Factorization for Private Training
• Multi-speaker Attention Alignment for Multimodal Social Interaction
• Mitigating Catastrophic Forgetting in Streaming Generative and Predictive Learning via Stateful Replay
• Modeling Retinal Ganglion Cells with Neural Differential Equations
• A Stitch in Time: Learning Procedural Workflow via Self-Supervised Plackett-Luce Ranking
• PrismSSL: One Interface, Many Modalities; A Single-Interface Library for Multimodal Self-Supervised Learning
• When Active Learning Fails, Uncalibrated Out of Distribution Uncertainty Quantification Might Be the Problem
• A First Full Physics Benchmark for Highly Granular Calorimeter Surrogates
• UI-Styler: Ultrasound Image Style Transfer with Class-Aware Prompts for Cross-Device Diagnosis Using a Frozen Black-Box Inference Network
• Real Noise Decoupling for Hyperspectral Image Denoising
• Fine-grained MoE Load Balancing with Linear Programming
• HGCN2SP: Hierarchical Graph Convolutional Network for Two-Stage Stochastic Programming
• Reinforcement learning of quantum circuit architectures for molecular potential energy curves
• Neural Positioning Without External Reference
• Beyond Tokens in Language Models: Interpreting Activations through Text Genre Chunks
• StreetView-Waste: A Multi-Task Dataset for Urban Waste Management
• VTinker: Guided Flow Upsampling and Texture Mapping for High-Resolution Video Frame Interpolation
• Warm-Starting Iterative Gaussian Processes for Faster Sequential Inference
• Memory-DD: A Low-Complexity Dendrite-Inspired Neuron for Temporal Prediction Tasks
• Target Refocusing via Attention Redistribution for Open-Vocabulary Semantic Segmentation: An Explainability Perspective
• How Noise Benefits AI-generated Image Detection
• Near-Field Topology-Optimized Superchiral Metasurfaces for Enhanced Chiral Sensing
• SpellForger: Prompting Custom Spell Properties In-Game using BERT supervised-trained model
• Hierarchical Semantic Tree Anchoring for CLIP-Based Class-Incremental Learning
• Real-Time Optimal Control via Transformer Networks and Bernstein Polynomials
• From Low-Rank Features to Encoding Mismatch: Rethinking Feature Distillation in Vision Transformers
• A Physics Informed Machine Learning Framework for Optimal Sensor Placement and Parameter Estimation
• DCC: Data-Centric Compilation of Machine Learning Kernels for Processing-In-Memory Architectures
• A Dataset and Baseline for Deep Learning-Based Visual Quality Inspection in Remanufacturing
• HV-Attack: Hierarchical Visual Attack for Multimodal Retrieval Augmented Generation
• jaxFMM: An Adaptive, GPU-Parallel Implementation of the Fast Multipole Method in JAX
• SNAP: Low-Latency Test-Time Adaptation with Sparse Updates
• Why Physics Still Matters: Improving Machine Learning Prediction of Material Properties with Phonon-Informed Datasets
• Insert In Style: A Zero-Shot Generative Framework for Harmonious Cross-Domain Object Composition
• UniSER: A Foundation Model for Unified Soft Effects Removal
• BrainRotViT: Transformer-ResNet Hybrid for Explainable Modeling of Brain Aging from 3D sMRI
• TiCAL:Typicality-Based Consistency-Aware Learning for Multimodal Emotion Recognition
• GPU-Initiated Networking for NCCL
• Complex-Valued 2D Gaussian Representation for Computer-Generated Holography
• CoroAMU: Unleashing Memory-Driven Coroutines through Latency-Aware Decoupled Operations
• Fine-tuning Pre-trained Audio Models for COVID-19 Detection: A Technical Report
• SilverTorch: A Unified Model-based System to Democratize Large-Scale Recommendation on GPUs
• Adapformer: Adaptive Channel Management for Multivariate Time Series Forecasting
• Gradient-descent methods for quantum detector tomography
• The PID Controller Strikes Back: Classical Controller Helps Mitigate Barren Plateaus in Noisy Variational Quantum Circuits
• DeepBlip: Estimating Conditional Average Treatment Effects Over Time
• Improved Convergence in Parameter-Agnostic Error Feedback through Momentum
• Accelerating Automatic Differentiation of Direct Form Digital Filters
• Free Lunch to Meet the Gap: Intermediate Domain Reconstruction for Cross-Domain Few-Shot Learning
• Gradient Flows of Potential Energies in the Geometry of Sinkhorn Divergences
• ArbESC+: Arabic Enhanced Edit Selection System Combination for Grammatical Error Correction Resolving conflict and improving system combination in Arabic GEC
• ELiC: Efficient LiDAR Geometry Compression via Cross-Bit-depth Feature Propagation and Bag-of-Encoders
• Training-free Detection of AI-generated images via Cropping Robustness
• Insight-A: Attribution-aware for Multimodal Misinformation Detection
• High Dynamic Range 3D Gaussian Splatting via Luminance-Chromaticity Decomposition
• Adaptive Multi-Scale Integration Unlocks Robust Cell Annotation in Histopathology Images
• Discovering Operational Patterns Using Image-Based Convolutional Clustering and Composite Evaluation: A Case Study in Foundry Melting Processes
• MMD-Thinker: Adaptive Multi-Dimensional Thinking for Multimodal Misinformation Detection
• Uncertainty-aware Physics-informed Neural Networks for Robust CARS-to-Raman Signal Reconstruction
• Protein Secondary Structure Prediction Using 3D Graphs and Relation-Aware Message Passing Transformers
• Semantics and Content Matter: Towards Multi-Prior Hierarchical Mamba for Image Deraining
• A Smart-Glasses for Emergency Medical Services via Multimodal Multitask Learning
• Dimension vs. Precision: A Comparative Analysis of Autoencoders and Quantization for Efficient Vector Retrieval on BEIR SciFact
• On the Information Processing of One-Dimensional Wasserstein Distances with Finite Samples
• Learned Adaptive Kernels for High-Fidelity Image Downscaling
• Improving the Generalisation of Learned Reconstruction Frameworks
• Efficiently Training A Flat Neural Network Before It has been Quantizated
• Cross-View Cross-Modal Unsupervised Domain Adaptation for Driver Monitoring System
• Towards Temporal Fusion Beyond the Field of View for Camera-based Semantic Scene Completion
• DPVO-QAT++: Heterogeneous QAT and CUDA Kernel Fusion for High-Performance Deep Patch Visual Odometry
• HiGFA: Hierarchical Guidance for Fine-grained Data Augmentation with Diffusion Models
• Optimal Self-Consistency for Efficient Reasoning with Large Language Models
• Optimising Density Computations in Probabilistic Programs via Automatic Loop Vectorisation
• A Disease-Aware Dual-Stage Framework for Chest X-ray Report Generation
• Combining Serverless and High-Performance Computing Paradigms to support ML Data-Intensive Applications
• TSGDiff: Rethinking Synthetic Time Series Generation from a Pure Graph Perspective
• Rethinking Multimodal Point Cloud Completion: A Completion-by-Correction Perspective
• Uncertainty-Guided Selective Adaptation Enables Cross-Platform Predictive Fluorescence Microscopy
• ReCast: Reliability-aware Codebook Assisted Lightweight Time Series Forecasting
• Sparse by Rule: Probability-Based N:M Pruning for Spiking Neural Networks
• Treatment Stitching with Schrödinger Bridge for Enhancing Offline Reinforcement Learning in Adaptive Treatment Strategies
• Mesh-based Super-resolution of Detonation Flows with Multiscale Graph Transformers
• An Adjoint Formulation of Energetic Particle Confinement
• On the Entropy Calibration of Language Models
• Learning the relative composition of EEG signals using pairwise relative shift pretraining
• Batch Matrix-form Equations and Implementation of Multilayer Perceptrons
• FreDN: Spectral Disentanglement for Time Series Forecasting via Learnable Frequency Decomposition
• Enhancing Photon Identification with Neural Network Methods
• BOFA: Bridge-Layer Orthogonal Low-Rank Fusion for CLIP-Based Class-Incremental Learning
• Log-Averaged Mirror Prox for Fast, Large-Scale Optimal Transport in Linear Space
• Dynamic Weight Adaptation in Spiking Neural Networks Inspired by Biological Homeostasis
• SurvBench: A Standardised Preprocessing Pipeline for Multi-Modal Electronic Health Record Survival Analysis
• Quantifying vacuum-like jets in heavy-ion collisions: a Machine Learning study
• ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference
• Know Your Limits: Entropy Estimation Modeling for Compression and Generalization
• STAGE: A Symbolic Tensor grAph GEnerator for distributed AI system co-design
• Learning parameter-dependent shear viscosity from data, with application to sea and land ice
• Torch-Uncertainty: A Deep Learning Framework for Uncertainty Quantification
• Data-driven multi-species heat flux closures for two-stream-unstable plasmas with nonlinear sparse regression
• DGFusion: Dual-guided Fusion for Robust Multi-Modal 3D Object Detection
• DreamPose3D: Hallucinative Diffusion with Prompt Learning for 3D Human Pose Estimation
• Distributional Shrinkage I: Universal Denoiser Beyond Tweedie's Formula
• LampQ: Towards Accurate Layer-wise Mixed Precision Quantization for Vision Transformers
• Harli: SLO-Aware Co-location of LLM Inference and PEFT-based Finetuning on Model-as-a-Service Platforms
• AdaptViG: Adaptive Vision GNN with Exponential Decay Gating
• EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models
• IPCD: Intrinsic Point-Cloud Decomposition
• Balancing Centralized Learning and Distributed Self-Organization: A Hybrid Model for Embodied Morphogenesis
• LLM Inference Beyond a Single Node: From Bottlenecks to Mitigations with Fast All-Reduce Communication
• An ultrafast plenoptic-camera system for high-resolution 3D particle tracking in unsegmented scintillators
• Free-Boundary Quasiconformal Maps via a Least-squares Operator in Diffeomorphism Optimization
• A Neural-Operator Preconditioned Newton Method for Accelerated Nonlinear Solvers
• Field theoretic atomistics: Learning thermodynamic and variational surrogate to density functional theory
• RESTL: Reinforcement Learning Guided by Multi-Aspect Rewards for Signal Temporal Logic Transformation
• MVSMamba: Multi-View Stereo with State Space Model
• A Spiking Sequence Generator for Polar Trajectories on Neuromorphic Hardware
• Event Tensor: A Unified Abstraction for Compiling Dynamic Megakernel
• coelsch: Platform-agnostic single-cell analysis of meiotic recombination events
• Generalizable Blood Cell Detection via Unified Dataset and Faster R-CNN
• One Model for All: Universal Pre-training for EEG based Emotion Recognition across Heterogeneous Datasets and Paradigms
• NeuCLIP: Efficient Large-Scale CLIP Training with Neural Normalizer Optimization
• Hybrid Quantum-Classical Selective State Space Artificial Intelligence
• NeuSpring: Neural Spring Fields for Reconstruction and Simulation of Deformable Objects from Videos
• Data-Driven Discovery of Feature Groups in Clinical Time Series
• From Classical to Hybrid: A Practical Framework for Quantum-Enhanced Learning
• KPLM-STA: Physically-Accurate Shadow Synthesis for Human Relighting via Keypoint-Based Light Modeling
• Generalizable Insights for Graph Transformers in Theory and Practice
• The Impact of Longitudinal Mammogram Alignment on Breast Cancer Risk Assessment
• Privacy Beyond Pixels: Latent Anonymization for Privacy-Preserving Video Understanding
• DI3CL: Contrastive Learning With Dynamic Instances and Contour Consistency for SAR Land-Cover Classification Foundation Model
• TurboSAT: Gradient-Guided Boolean Satisfiability Accelerated on GPU-CPU Hybrid System
• Quantum-centric machine learning for molecular dynamics
• Enhancing Binary Encoded Crime Linkage Analysis Using Siamese Network
• AutoPollS: A tool for automated monitoring of pollinators using deep learning
• Bridging taxonomic gaps in microbial community profiling with LSTM-generated synthetic full-length 16S rRNA sequences
• Lightning Grasp: High Performance Procedural Grasp Synthesis with Contact Fields
• LoReTTA: A Low Resource Framework To Poison Continuous Time Dynamic Graphs
• AcousTools: A 'Full-Stack', Python-Based, Acoustic Holography Library
• LoRA on the Go: Instance-level Dynamic LoRA Selection and Merging
• Cross-Modal Unlearning via Influential Neuron Path Editing in Multimodal Large Language Models
• Robust and High-Fidelity 3D Gaussian Splatting: Fusing Pose Priors and Geometry Constraints for Texture-Deficient Outdoor Scenes
• Multi-Modal Continual Learning via Cross-Modality Adapters and Representation Alignment with Knowledge Preservation
• MirrorMamba: Towards Scalable and Robust Mirror Detection in Videos
• Speech Separation for Hearing-Impaired Children in the Classroom
• Magnitude-Modulated Equivariant Adapter for Parameter-Efficient Fine-Tuning of Equivariant Graph Neural Networks
• Fast Riemannian-manifold Hamiltonian Monte Carlo for hierarchical Gaussian-process models
• Reaction Prediction via Interaction Modeling of Symmetric Difference Shingle Sets
• Improving Multimodal Sentiment Analysis via Modality Optimization and Dynamic Primary Modality Selection
• Robust Differentiable Collision Detection for General Objects
• Functional Adjoint Sampler: Scalable Sampling on Infinite Dimensional Spaces
• Optimizing Long-context LLM Serving via Fine-grained Sequence Parallelism
• Zero-Shot Function Encoder-Based Differentiable Predictive Control
• Adversarially Regularized Policy Learning Guided by Trajectory Optimization
• Entropy-Rank Ratio: A Novel Entropy-Based Perspective for DNA Complexity and Classification
• QuAnTS: Question Answering on Time Series
• Self-adaptive weighting and sampling for physics-informed neural networks
• IndicVisionBench: Benchmarking Cultural and Multilingual Understanding in VLMs
• wa-hls4ml: A Benchmark and Surrogate Models for hls4ml Resource and Latency Estimation
• Probabilistic Textual Time Series Depression Detection
• Landslide Hazard Mapping with Geospatial Foundation Models: Geographical Generalizability, Data Scarcity, and Band Adaptability
• GRAPE.jl: Gradient Ascent Pulse Engineering in Julia
• Guided by Stars: Interpretable Concept Learning Over Time Series via Temporal Logic Semantics
• Accelerated Sequential Posterior Inference via Reuse for Gravitational-Wave Analyses
• Seeing Straight: Document Orientation Detection for Efficient OCR
• Tortoise and Hare Guidance: Accelerating Diffusion Model Inference with Multirate Integration
• CBMC-V3: A CNS-inspired Control Framework Towards Manipulation Agility with SNN
• When Swin Transformer Meets KANs: An Improved Transformer Architecture for Medical Image Segmentation
• MedDChest: A Content-Aware Multimodal Foundational Vision Model for Thoracic Imaging
• Validating a Machine Learning Approach to Identify Quenched Jets in Heavy-Ion Collisions
• Structured Matrix Scaling for Multi-Class Calibration
• Robust electron counting for direct electron detectors with the Back-Propagation Counting method
• Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs
• Part-Aware Bottom-Up Group Reasoning for Fine-Grained Social Interaction Detection
• Signal Intensity-weighted coordinate channels improve learning stability and generalisation in 1D and 2D CNNs in localisation tasks on biomedical signals
• Bearing Syntactic Fruit with Stack-Augmented Neural Networks
• Hybrid DeepONet Surrogates for Multiphase Flow in Porous Media
• Test-Time Steering for Lossless Text Compression via Weighted Product of Experts
• SEAL - A Symmetry EncourAging Loss for High Energy Physics
• Unsupervised Learning for Industrial Defect Detection: A Case Study on Shearographic Data
• Keeping it Local, Tiny and Real: Automated Report Generation on Edge Computing Devices for Mechatronic-Based Cognitive Systems
• From the Laboratory to Real-World Application: Evaluating Zero-Shot Scene Interpretation on Edge Devices for Mobile Robotics
• Self-Supervised Moving Object Segmentation of Sparse and Noisy Radar Point Clouds
• GAFD-CC: Global-Aware Feature Decoupling with Confidence Calibration for OOD Detection
• DL4Proteins Jupyter Notebooks Teach how to use Artificial Intelligence for Biomolecular Structure Prediction and Design
• TACO: Trajectory-Aware Controller Optimization for Quadrotors
• Predicting Microbial Interactions Using Graph Neural Networks
• LM-Fix: Lightweight Bit-Flip Detection and Rapid Recovery Framework for Language Models
• Design, Assessment, and Application of Machine Learning Potential Energy Surfaces
• Assessing LLM Reasoning Steps via Principal Knowledge Grounding
• Investigating the Robustness of Knowledge Tracing Models in the Presence of Student Concept Drift
• EPARA: Parallelizing Categorized AI Inference in Edge Clouds
• Digital Twin of Aerosol Jet Printing
• Trust-Region Methods with Low-Fidelity Objective Models
• Saliency-R1: Incentivizing Unified Saliency Reasoning Capability in MLLM with Confidence-Guided Reinforcement Learning
• Superpositional Gradient Descent: Harnessing Quantum Principles for Model Training
• Sensitivity Analysis for Climate Science with Generative Flow Models
• Variational Geometry-aware Neural Network based Method for Solving High-dimensional Diffeomorphic Mapping Problems
• PETAR: Localized Findings Generation with Mask-Aware Vision-Language Modeling for PET Automated Reporting
• Vectorized Online POMDP Planning
• Cognitive Alignment in Personality Reasoning: Leveraging Prototype Theory for MBTI Inference
• pDANSE: Particle-based Data-driven Nonlinear State Estimation from Nonlinear Measurements
• LongCat-Flash-Omni Technical Report
• SU(N) lattice gauge theories with Physics-Informed Neural Networks
• Defeating the Training-Inference Mismatch via FP16
• LLMs Process Lists With General Filter Heads
• FlowQ-Net: A Generative Framework for Automated Quantum Circuit Design
• MSAD: A Deep Dive into Model Selection for Time series Anomaly Detection
• Analysis of the Robustness of an Edge Detector Based on Cellular Automata Optimized by Particle Swarm
• Robust variable selection for spatial point processes observed with noise
• Similarity-Distance-Magnitude Language Models
• Group-Equivariant Diffusion Models for Lattice Field Theory
• Data-driven Projection Generation for Efficiently Solving Heterogeneous Quadratic Programming Problems
• Detecting Anomalies in Machine Learning Infrastructure via Hardware Telemetry
• Active Learning with Task-Driven Representations for Messy Pools
• Evaluation of Wafer-Scale SOT-MRAM for Analog Crossbar Array Applications
• PyDPF: A Python Package for Differentiable Particle Filtering
• Demystifying Deep Learning Compiler Frontend Bugs: An LLM-Aided Empirical Study
• DynaBridge: Dynamic Summary-Guided Cross-Task Multimodal Fusion for DASS-Structured Mental Health Assessment
• Gestalt: a Stacking Ensemble for SQuAD2.0
• Target Propagation via Regularized Inversion
• Exemplar-Based Open-Set Panoptic Segmentation Network
• Competing AI: How does competition feedback affect machine learning?
• Deep Multi-Fidelity Active Learning of High-dimensional Outputs
• Decentralized Reinforcement Learning: Global Decision-Making via Local Economic Transactions
• Self-Regulation for Semantic Segmentation
• Efficient Attentions for Long Document Summarization
• Deep learning models for unbiased sequence-based PPI prediction plateau at an accuracy of 0.65
• Bayesian reaction optimization as a tool for chemical synthesis
• Molecular optimization using a conditional transformer for reaction-aware compound exploration with reinforcement learning
• Achieving Real-Time LiDAR 3D Object Detection on a Mobile Device
• Bayesian Bits: Unifying Quantization and Pruning
• ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations
• DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing
• Interpretable Image-Level Acne Severity Grading via EfficientNet-B0 Transfer Learning and Grad-CAM
• StrataCL: Fabric-Native Communication Library for Production Supernodes
• CASIAL: Geometric Distortion Robust Image Watermarking
• JEPADepth: Masked Predictive Representation Learning for Self-Supervised Monocular Depth Estimation
• Learning the Word Problem: Geodesic Lengths and Cryptographic Applications
• DVPSFormer: Efficient Online Depth-aware Video Panoptic Segmentation for Autonomous Driving
• The Confounder Trap: Treatment-Encoding Representations in Causal Inference with Text
• Mixture-of-Depths Attention
• TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning
• Multi-Class Multiple Instance Learning for Predicting Precursors to\n Aviation Safety Events
• A Tale of Two Linkings: Dynamically Gating between Schema Linking and Structural Linking for Text-to-SQL Parsing
• The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Models
• Rotational equivariance and locality in data-driven subgrid-scale closures
• Real-time nonlinear inversion of magnetic resonance elastography with operator learning
• Learning Backward Transport for Source Localization
• Uncovering the structure of clinical EEG signals with self-supervised\n learning
• QPIC: Query-Based Pairwise Human-Object Interaction Detection with Image-Wide Contextual Information
• Vehicle tracks in the <scp>UK</scp> uplands vary in occurrence with land use and overlap with peatlands and protected areas
• MisConv: Convolutional Neural Networks for Missing Data
• Advancing marine microplastic monitoring through deep learning-based image segmentation
• Deep Structural Causal Models for Tractable Counterfactual Inference
• Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI
• TweetyBERT: Automated parsing of birdsong through self-supervised machine learning
• Lightning IR: Straightforward Fine-tuning and Inference of Transformer-based Language Models for Information Retrieval
• Feature Extraction for Novelty Detection in Network Traffic
• Characterizing Real-World Bugs in Tile Programs for Automated Bug Detection
• AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation
• Let Your Heart Speak in its Mother Tongue: Multilingual Captioning of Cardiac Signals
• FedLab: A Flexible Federated Learning Framework
• Temporal Spike Sequence Learning via Backpropagation for Deep Spiking Neural Networks
• Heavy Tails in SGD and Compressibility of Overparametrized Neural Networks
• Array programming with NumPy
• Dynamic Resolution Network
• Self-Supervised Monocular Depth Estimation with Internal Feature Fusion
• DeepACEv2: Automated Chromosome Enumeration in Metaphase Cell Images Using Deep Convolutional Neural Networks
• MIND: Monge Inception Distance for Generative Models Evaluation
• DSA: More Efficient Budgeted Pruning via Differentiable Sparsity Allocation
• MixPath: A Unified Approach for One-shot Neural Architecture Search
• A Meta-Learning Approach for Graph Representation Learning in Multi-Task Settings
• Co-Imitation Learning without Expert Demonstration
• Measuring Dependence with Matrix-based Entropy Functional
• Coordinate Attention for Efficient Mobile Network Design
• CompressAI: a PyTorch library and evaluation platform for end-to-end\n compression research
• LGESQL: Line Graph Enhanced Text-to-SQL Model with Mixed Local and Non-Local Relations
• Boosting Salient Object Detection with Transformer-based Asymmetric Bilateral U-Net
• FunMC: A functional API for building Markov Chains
• DIVeR: Real-time and Accurate Neural Radiance Fields with Deterministic Integration for Volume Rendering
• EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback
• Efficient Spatialtemporal Context Modeling for Action Recognition
• The Yeast Lifespan Machine: a microfluidic platform for automated replicative lifespan measurements
• Effect of the output activation function on the probabilities and errors in medical image segmentation
• Balanced Meta-Softmax for Long-Tailed Visual Recognition
• AANet: Adaptive Aggregation Network for Efficient Stereo Matching
• Deep Adaptive Design: Amortizing Sequential Bayesian Experimental Design
• Detection of Consonant Errors in Disordered Speech Based on Consonant-vowel Segment Embedding
• SiMaN: Sign-to-Magnitude Network Binarization
• Bone Feature Segmentation in Ultrasound Spine Image with Robustness to Speckle and Regular Occlusion Noise
• VisCode: Embedding Information in Visualization Images using Encoder-Decoder Network
• IntraQ: Learning Synthetic Images with Intra-Class Heterogeneity for Zero-Shot Network Quantization
• PAGE: A Simple and Optimal Probabilistic Gradient Estimator for Nonconvex Optimization
• CLDA: Contrastive Learning for Semi-Supervised Domain Adaptation
• Low-latency Event-based Object Detection with Spatially-Sparse Linear Attention
• A Fourier-Space Approach to Physics-Informed Magnetization Reconstruction from Nitrogen-Vacancy Measurements
• Picasso: Holistic Scene Reconstruction with Physics-Constrained Sampling
• Symmetry and Generalisation in Neural Approximations of Renormalisation Transformations
• AI-Based Regional Emulation for Kilometer-Scale Dynamical Downscaling
• genriesz: A Python Package for Automatic Debiased Machine Learning with Generalized Riesz Regression
• RaCo: Ranking and Covariance for Practical Learned Keypoints
• Trex: Learning Execution Semantics from Micro-Traces for Binary Similarity
• Predicting life satisfaction using machine learning and explainable AI
• Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
• Cycle Self-Training for Domain Adaptation
• GeoAI: A Python package for integrating artificial intelligence with geospatial data analysis and visualization
• A robust and stable hybrid neural network/finite element method for 2D flows that generalizes to different geometries
• Knowledge Distillation of a Protein Language Model Yields a Foundational Implicit Solvent Model
• Regime-Adaptive Bayesian Optimization via Dirichlet Process Mixtures of Gaussian Processes
• A Distributed Multi-GPU System for Large-Scale Node Embedding at Tencent
• MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Images
• SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure
• Certifiable Machine Unlearning for Linear Models
• Involution: Inverting the Inherence of Convolution for Visual Recognition
• Guiding questions to avoid data leakage in biological machine learning applications
• Maximum Likelihood Estimation for Multimodal Learning with Missing Modality
• Accelerating GMRES with Deep Learning in Real-Time
• Multi-Resolution Model Fusion for Accelerating the Convolutional Neural Network Training
• Learning Disentangled Speech- and Expression-Driven Blendshapes for 3D Talking Face Animation
• Fast chaos indicator from auto-differentiation for dynamic aperture optimization
• Adversarially Robust Quantum Transfer Learning
• Cosmic Background Removal with Deep Neural Networks in SBND
• Coordinate-wise Control Variates for Deep Policy Gradients
• Region-CAM: Towards Accurate Object Regions in Class Activation Maps for Weakly Supervised Learning Tasks
• AtlasGS: Atlanta-world Guided Surface Reconstruction with Implicit Structured Gaussians
• Deep Deterministic Information Bottleneck with Matrix-based Entropy Functional
• A Quadratic Actor Network for Model-Free Reinforcement Learning
• Emergence of Minimal Circuits for Indirect Object Identification in Attention-Only Transformers
• FaRAccel: FPGA-Accelerated Defense Architecture for Efficient Bit-Flip Attack Resilience in Transformer Models
• SemCoT: Accelerating Chain-of-Thought Reasoning through Semantically-Aligned Implicit Tokens
• SHA-256 Infused Embedding-Driven Generative Modeling of High-Energy Molecules in Low-Data Regimes
• TRIE: End-to-End Text Reading and Information Extraction for Document Understanding
• E-DSSR: Efficient Dynamic Surgical Scene Reconstruction with Transformer-based Stereoscopic Depth Perception
• Unsupervised Machine-Learning Pipeline for Data-Driven Defect Detection and Characterisation: Application to Displacement Cascades
• Detecting dementia from speech and transcripts using transformers
• Synergizing chemical and AI communities for advancing laboratories of the future
• MAGNET: A Multi-Graph Attentional Network for Code Clone Detection
• Causal Convolutional Neural Networks as Finite Impulse Response Filters
• MC-SJD : Maximal Coupling Speculative Jacobi Decoding for Autoregressive Visual Generation Acceleration
• Beyond Neural Incompatibility: Easing Cross-Scale Knowledge Transfer in Large Language Models through Latent Semantic Alignment
• EddyFormer: Accelerated Neural Simulations of Three-Dimensional Turbulence at Scale
• Beyond Line-Level Filtering for the Pretraining Corpora of LLMs
• Adaptive Training of INRs via Pruning and Densification
• Scalable GPU-Based Integrity Verification for Large Machine Learning Models
• Enhancing Pre-trained Representation Classifiability can Boost its Interpretability
• Deep Learning-Enhanced Calibration of the Heston Model: A Unified Framework
• Taming Visually Guided Sound Generation
• GraphNet: A Large-Scale Computational Graph Dataset for Tensor Compiler Research
• Key and Value Weights Are Probably All You Need: On the Necessity of the Query, Key, Value weight Triplet in Decoder-Only Transformers
• MoPHES:Leveraging on-device LLMs as Agent for Mobile Psychological Health Evaluation and Support
• From Subconscious to Insight: Decoding the Incubation Process in Creative Problem Solving
• Benchmarking Simulation-Based Inference
• Multi-Task Reinforcement Learning with Context-based Representations
• Fast and scalable joint-LORAKS reconstruction and data-driven sampling optimisation of high-dimensional MRI datasets using a GPU-accelerated and learning-free differentiable framework: PyLORAKS
• Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction
• Strengths and Limitations of Statistical and Dynamical Downscaling for the Representation of Compound Dry and Hot Events Over Spain
• Behavioral alignment as an organizing principle in sensory coding
• Implicit recurrent networks: A novel approach to stationary input processing with recurrent neural networks in deep learning
• PySCIPOpt-ML: Embedding Trained Machine Learning Models into Mixed-Integer Programs
• Super-recognizers sample visual information of superior computational value for facial recognition
• ScaLoRA: Optimally Scaled Low-Rank Adaptation for Efficient High-Rank Fine-Tuning
• VIPAMIN: Visual Prompt Initialization via Embedding Selection and Subspace Expansion
• Deep learning–based frameworks for the detection and classification of soniferous fish
• Maximal Load Shedding Verification for Neural Network Models of AC Line Switching
• IoU-aware Single-stage Object Detector for Accurate Localization
• Sequence Modeling with Spectral Mean Flows
• MergeMix: A Unified Augmentation Paradigm for Visual and Multi-Modal Understanding
• High-resolution Piano Transcription with Pedals by Regressing Onset and Offset Times
• Initialization and Regularization of Factorized Neural Layers
• A Pragmatic Look at Deep Imitation Learning
• 3rd Place Scheme on Instance Segmentation Track of ICCV 2021 VIPriors Challenges
• An Efficient Remote Sensing Super Resolution Method Exploring Diffusion Priors and Multi-Modal Constraints for Crop Type Mapping
• Model-Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn't the Right One
• Friction on Demand: A Generative Framework for the Inverse Design of Metainterfaces
• Process Reward Models for Sentence-Level Verification of LVLM Radiology Reports
• The 2020 ESPnet update: new features, broadened applications, performance improvements, and future plans
• Implicit Modeling for Transferability Estimation of Vision Foundation Models
• Residual Diffusion Bridge Model for Image Restoration
• How Muon's Spectral Design Benefits Generalization: A Study on Imbalanced Data
• Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
• Clustering by Denoising: Latent plug-and-play diffusion for single-cell data
• Efficient Person Search: An Anchor-Free Approach
• Transformers from Compressed Representations
• Learning Without Augmenting: Unsupervised Time Series Representation Learning via Frame Projections
• Enhancing Graph Classification Robustness with Singular Pooling
• Statistical Analysis of the Sinkhorn Iterations for Two-Sample Schrödinger Bridge Estimation
• On the Iteration Complexity of Hypergradient Computation
• EventHPE: Event-based 3D Human Pose and Shape Estimation
• An Interval Hessian-based line-search method for unconstrained nonconvex optimization
• Empowering Multimodal Respiratory Sound Classification with Counterfactual Adversarial Debiasing for Out-of-Distribution Robustness
• GALA: A GlobAl-LocAl Approach for Multi-Source Active Domain Adaptation
• Simplifying Knowledge Transfer in Pretrained Models
• An Improved Analysis of Stochastic Gradient Descent with Momentum
• Tractable Shapley Values and Interactions via Tensor Networks
• ReDet: A Rotation-equivariant Detector for Aerial Object Detection
• The Power of Log-Sum-Exp: Sequential Density Ratio Matrix Estimation for Speed-Accuracy Optimization
• Revisiting Orbital Minimization Method for Neural Operator Decomposition
• Out-of-Scope Intent Detection with Self-Supervision and Discriminative Training
• A Multimodal Benchmark for Framing of Oil &amp; Gas Advertising and Potential Greenwashing Detection
• Weak-to-Strong Generalization under Distribution Shifts
• Chronos-2: From Univariate to Universal Forecasting
• Relieving the Over-Aggregating Effect in Graph Transformers
• Unified Implementations of Recurrent Neural Networks in Multiple Deep Learning Frameworks
• Quantum Neural Network Architectures for Multivariate Time-Series Forecasting
• How to Auto-optimize Prompts for Domain Tasks? Adaptive Prompting and Reasoning through Evolutionary Domain Knowledge Adaptation
• Deep Graph Contrastive Representation Learning
• SAMix: Calibrated and Accurate Continual Learning via Sphere-Adaptive Mixup and Neural Collapse
• Semi-Supervised Crowd Counting via Self-Training on Surrogate Tasks
• Self-Rewarding PPO: Aligning Large Language Models with Demonstrations Only
• Long-tailed Species Recognition in the NACTI Wildlife Dataset
• xMem: A CPU-Based Approach for Accurate Estimation of GPU Memory in Deep Learning Training Workloads
• Attention Sinks in Diffusion Language Models
• Gaussian Mixture Flow Matching with Domain Alignment for Multi-Domain Sequential Recommendation
• MirrorFuzz: Leveraging LLM and Shared Bugs for Deep Learning Framework APIs Fuzzing
• Pty-Chi: A PyTorch-based modern ptychographic data analysis package
• Preventing Shortcuts in Adapter Training via Providing the Shortcuts
• AutoScape: Geometry-Consistent Long-Horizon Scene Generation
• A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks
• Why Did Apple Fall To The Ground: Evaluating Curiosity In Large Language Model
• H-SPLID: HSIC-based Saliency Preserving Latent Information Decomposition
• Subject-Aware Contrastive Learning for Biosignals
• Bayesian Inference for PDE-based Inverse Problems using the Optimization of a Discrete Loss
• Evidential Turing Processes
• Predicting the 3D microstructure of SOFC anodes from 2D SEM images using stochastic microstructure modeling and CNNs
• Revealing systematics in phenomenologically viable flux vacua with reinforcement learning
• Addressing Mark Imbalance in Integration-free Neural Marked Temporal Point Processes
• Inverse Image-Based Rendering for Light Field Generation from Single Images
• Why Prototypes Collapse: Diagnosing and Preventing Partial Collapse in Prototypical Self-Supervised Learning
• Extending machine learning model for implicit solvation to free energy calculations
• Structural Invariance Matters: Rethinking Graph Rewiring through Graph Metrics
• Integrating Machine Learning into Belief-Desire-Intention Agents: Current Advances and Open Challenges
• KL-Regularized Reinforcement Learning is Designed to Mode Collapse
• GMFVAD: Using Grained Multi-modal Feature to Improve Video Anomaly Detection
• Kinetics of Peierls dimerization transition: Machine learning force-field approach
• Predicting before Reconstruction: A generative prior framework for MRI acceleration
• HybridEP: Scaling Expert Parallelism to Cross-Datacenter Scenario via Hybrid Expert/Data Transmission
• Neuromorphic intermediate representation: A unified instruction set for interoperable brain-inspired computing
• A Study of Face Obfuscation in ImageNet
• ActNN: Reducing Training Memory Footprint via 2-Bit Activation Compressed Training
• Continual Speaker Adaptation for Text-to-Speech Synthesis
• AI Pose Analysis and Kinematic Profiling of Range-of-Motion Variations in Resistance Training
• Environment Inference for Learning Generalizable Dynamical System
• Demonstrating Real Advantage of Machine-Learning-Enhanced Monte Carlo for Combinatorial Optimization
• No RL, No Simulation: Learning to Navigate without Navigating
• CoRECT: A Framework for Evaluating Embedding Compression Techniques at Scale
• Pointwise Binary Classification with Pairwise Confidence Comparisons
• Optimization Benchmark for Diffusion Models on Dynamical Systems
• Automated Concern Extraction from Textual Requirements of Cyber-Physical Systems: A Multi-solution Study
• Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID
• CDI-DTI: A Strong Cross-domain Interpretable Drug-Target Interaction Prediction Framework Based on Multi-Strategy Fusion
• UniHPR: Unified Human Pose Representation via Singular Value Contrastive Learning
• Weight Decay may matter more than muP for Learning Rate Transfer in Practice
• An Encode-then-Decompose Approach to Unsupervised Time Series Anomaly Detection on Contaminated Training Data--Extended Version
• A Unified Perspective on Optimization in Machine Learning and Neuroscience: From Gradient Descent to Neural Adaptation
• CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training
• On-Device Inference versus Wireless Streaming: Energy-Efficient Multi-Modal Deep Learning for Wearable Cardiovascular Patches
• TSception: A Deep Learning Framework for Emotion Detection Using EEG
• Overinterpretation reveals image classification model pathologies
• NeRD: Neural Representation of Distribution for Medical Image Segmentation
• Learning landmark geodesics using Kalman ensembles
• CytoNet: A Foundation Model for the Human Cerebral Cortex at Cellular Resolution
• Mono4DGS-HDR: High Dynamic Range 4D Gaussian Splatting from Alternating-exposure Monocular Videos
• Learning Boltzmann Generators via Constrained Mass Transport
• DiffGRM: Diffusion-based Generative Recommendation Model
• Towards Robust Zero-Shot Reinforcement Learning
• A Neural-Mean Vecchia Gaussian Process for Unified Argo Modeling
• Differentiable Model Compression via Pseudo Quantization Noise
• Is Multilingual LLM Watermarking Truly Multilingual? A Simple Back-Translation Solution
• Paying Attention to Activation Maps in Camera Pose Regression
• DeepDetect: Learning All-in-One Dense Keypoints
• HittER: Hierarchical Transformers for Knowledge Graph Embeddings
• Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads
• Back to Bytes: Revisiting Tokenization Through UTF-8
• All You Need is One: Capsule Prompt Tuning with a Single Vector
• Mixed-Precision Quantization for Language Models: Techniques and Prospects
• Adversarial Reinforcement Learning for Robust Control of Fixed-Wing Aircraft under Model Uncertainty
• Geospatial Machine Learning Libraries
• Stacked Temporal Attention: Improving First-person Action Recognition by Emphasizing Discriminative Clips
• Syntax Customized Video Captioning by Imitating Exemplar Sentences
• Quantile Filtered Imitation Learning
• Self-Certifying Primal-Dual Optimization Proxies for Large-Scale Batch Economic Dispatch
• Compressive Modeling and Visualization of Multivariate Scientific Data using Implicit Neural Representation
• PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks
• Extending Audio Context for Long-Form Understanding in Large Audio-Language Models
• Few-Shot Drum Transcription in Polyphonic Music
• XLVIN: eXecuted Latent Value Iteration Nets
• A Geometric Approach to Optimal Experimental Design
• OCR-APT: Reconstructing APT Stories from Audit Logs using Subgraph Anomaly Detection and LLMs
• Exploring the Synergy of Quantitative Factors and Newsflow Representations from Large Language Models for Stock Return Prediction
• Hyperparameter Optimization and Reproducibility in Deep Learning Model Training
• Extending Temporal Disturbance Estimations For Magnetic Anomaly Navigation and Mapping
• Learning invariance preserving moment closure model for Boltzmann-BGK equation
• Decorrelation Speeds Up Vision Transformers
• SaLon3R: Structure-aware Long-term Generalizable 3D Reconstruction from Unposed Images
• A solution to generalized learning from small training sets found in infant repeated visual experiences of individual objects
• Biology-informed neural networks learn nonlinear representations from omics data to improve genomic prediction and interpretability
• Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References
• On The Expressive Power of GNN Derivatives
• Enhancing Time Series Forecasting through Selective Representation Spaces: A Patch Perspective
• cubic: CUDA-accelerated 3D Bioimage Computing
• Synchronization of Multiple Videos
• Hyper-Parameter Optimization: A Review of Algorithms and Applications
• DeepHunter: A Graph Neural Network Based Approach for Robust Cyber Threat Hunting
• Tensor Gaussian Processes: Efficient Solvers for Nonlinear PDEs
• Message Passing on the Edge: Towards Scalable and Expressive GNNs
• A General Family of Stochastic Proximal Gradient Methods for Deep Learning
• In-Distribution Steering: Balancing Control and Coherence in Language Model Generation
• Prompt-based Adaptation in Large-scale Vision Models: A Survey
• AGNES: Adaptive Graph Neural Network and Dynamic Programming Hybrid Framework for Real-Time Nanopore Seed Chaining
• Hybrid Boson Sampling-Neural Network Architecture for Enhanced Classification
• Multi-Scale High-Resolution Logarithmic Grapher Module for Efficient Vision GNNs
• Repairing Reward Functions with Human Feedback to Mitigate Reward Hacking
• A Neural Network Ensemble Approach to System Identification
• A Connection Between Score Matching and Local Intrinsic Dimension
• SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms
• EfficientPhys: Enabling Simple, Fast and Accurate Camera-Based Vitals Measurement
• Learning to Grasp Anything by Playing with Random Toys
• Efficient Perceptual Image Super Resolution: AIM 2025 Study and Benchmark
• AI-Assisted Physics-Informed Predictions of Degradation Behavior of Polymeric Anion Exchange Membranes
• PromoGuardian: Detecting Promotion Abuse Fraud with Multi-Relation Fused Graph Neural Networks
• ProtoSiTex: Learning Semi-Interpretable Prototypes for Multi-label Text Classification
• A Function Centric Perspective On Flat and Sharp Minima
• Search for dark matter production in association with bottom quarks and a lepton pair in proton-proton collisions at √(s) = 13 TeV
• G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior
• PyTorch Tabular: A Framework for Deep Learning with Tabular Data
• Deeplite Neutrino: An End-to-End Framework for Constrained Deep Learning Model Optimization
• Multi-Action Self-Improvement for Neural Combinatorial Optimization
• Conditional Positional Encodings for Vision Transformers
• Relative Flatness and Generalization
• The Ivory Tower Lost: How College Students Respond Differently than the General Public to the COVID-19 Pandemic
• Multi-Graph Transformer for Free-Hand Sketch Recognition
• Enhancing Pre-trained Chinese Character Representation with Word-aligned Attention
• ANCER: Anisotropic Certification via Sample-wise Volume Maximization
• Active Subspaces in Infinite Dimension
• PhySIC: Physically Plausible 3D Human-Scene Interaction and Contact from a Single Image
• Enforcing convex constraints in Graph Neural Networks
• NeuralProphet: Explainable Forecasting at Scale
• ROFI: A Deep Learning-Based Ophthalmic Sign-Preserving and Reversible Patient Face Anonymizer
• A Constrained Multi-Fidelity Bayesian Optimization Method
• In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
• More than A Point: Capturing Uncertainty with Adaptive Affordance Heatmaps for Spatial Grounding in Robotic Tasks
• Learning the syntax of plant assemblages
• TorchCor: High-Performance Cardiac Electrophysiology Simulations with the Finite Element Method on GPUs
• Wasserstein normalized autoencoder for anomaly detection
• Digital Twin-enabled Multi-generation Control Co-Design with Deep Reinforcement Learning
• Learning from Disagreement: A Group Decision Simulation Framework for Robust Medical Image Segmentation
• PENEX: AdaBoost-Inspired Neural Network Regularization
• DEMO: Disentangled Motion Latent Flow Matching for Fine-Grained Controllable Talking Portrait Synthesis
• CryptGPU: Fast Privacy-Preserving Machine Learning on the GPU
• The Achilles' Heel of LLMs: How Altering a Handful of Neurons Can Cripple Language Abilities
• Average Kernel Sizes -- Computable Sharp Accuracy Bounds for Inverse Problems
• Path Sample-Analytic Gradient Estimators for Stochastic Binary Networks
• AP20-OLR Challenge: Three Tasks and Their Baselines
• Peransformer: Improving Low-informed Expressive Performance Rendering with Score-aware Discriminator
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting
• PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models
• Preference-driven Knowledge Distillation for Few-shot Node Classification
• Uncertainty-Aware Post-Detection Framework for Enhanced Fire and Smoke Detection in Compact Deep Learning Models
• Cooperative Pseudo Labeling for Unsupervised Federated Classification
• ARCH: Efficient Adversarial Regularized Training with Caching
• Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach
• Accelerating kinetic plasma simulations with machine learning generated initial conditions
• Flow-Matching Guided Deep Unfolding for Hyperspectral Image Reconstruction
• Behind Python: The Languages That Power AI
• Irrationality as a mean of regularization in Bayesian Persuasion
• Impact of Scanner Manufacturer, Endorectal Coil Use, and Clinical Variables on Deep Learning–assisted Prostate Cancer Classification Using Multiparametric MRI
• On the Quantization Robustness of Diffusion Language Models in Coding Benchmarks
• A Physical Theory of Backpropagation: Exact Gradients from the Least-Action Principle
• Condenser: a Pre-training Architecture for Dense Retrieval
• SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking
• HAFLO: GPU-Based Acceleration for Federated Logistic Regression
• Deep Learning-based Stress Determinator for Mouse Psychiatric Analysis using Hippocampus Activity
• The Blessing and the Curse of the Noise behind Facial Landmark Annotations
• Efficient Emulation of Neutral Atom Quantum Hardware
• StereoPIFu: Depth Aware Clothed Human Digitization via Stereo Vision
• Video Abnormal Event Detection by Learning to Complete Visual Cloze Tests
• Obstacle Avoidance using Dynamic Movement Primitives and Reinforcement Learning
• Maple: A Multi-agent System for Portable Deep Learning across Clusters
• 3D Reconstruction from Transient Measurements with Time-Resolved Transformer
• Denoised Diffusion for Object-Focused Image Augmentation
• Model-Based Lookahead Reinforcement Learning for in-hand manipulation
• MPA-DNN: Projection-Aware Unsupervised Learning for Multi-period DC-OPF
• MIP-Based Tumor Segmentation: A Radiologist-Inspired Approach
• Neural Codecs as Biosignal Tokenizers
• Application of Deep Reinforcement Learning to At-the-Money S&amp;P 500 Options Hedging
• Conformal Risk Training: End-to-End Optimization of Conformal Risk Control
• Neptune: Advanced ML Operator Fusion for Locality and Parallelism on GPUs
• ReSplat: Learning Recurrent Gaussian Splats
• DYNAMIX: RL-based Adaptive Batch Size Optimization in Distributed Machine Learning Systems
• AI-Driven Radiology Report Generation for Traumatic Brain Injuries
• Counterfactual Identifiability via Dynamic Optimal Transport
• Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction
• Efficient Neural Network Training via Forward and Backward Propagation\n Sparsification
• Conditional Alignment and Uniformity for Contrastive Learning with\n Continuous Proxy Labels
• Learning Deep Features in Instrumental Variable Regression
• Multilingual Generative Retrieval via Cross-lingual Semantic Compression
• Dual-Stream Alignment for Action Segmentation
• MixerGAN: An MLP-Based Architecture for Unpaired Image-to-Image Translation
• PhysCap: Physically Plausible Monocular 3D Motion Capture in Real Time
• A Distributed Training Algorithm of Generative Adversarial Networks with Quantized Gradients
• OBCache: Optimal Brain KV Cache Pruning for Efficient Long-Context LLM Inference
• Safely Exploring Novel Actions in Recommender Systems via Deployment-Efficient Policy Learning
• Characterizing the Value of Information in Medical Notes
• Nonparametric Causal Feature Selection for Spatiotemporal Risk Mapping of Malaria Incidence in Madagascar
• TextHide: Tackling Data Privacy in Language Understanding Tasks
• What Do You See? Evaluation of Explainable Artificial Intelligence (XAI) Interpretability through Neural Backdoors
• Cough Against COVID: Evidence of COVID-19 Signature in Cough Sounds
• SLGAN: Style- and Latent-guided Generative Adversarial Network for Desirable Makeup Transfer and Removal
• Rethinking Provenance Completeness with a Learning-Based Linux Scheduler
• Counterfactually Fair Conformal Prediction
• From Data to Rewards: a Bilevel Optimization Perspective on Maximum Likelihood Estimation
• Single and Multi-Objective Optimization of Distributed Acoustic Sensing Cable Layouts for Geophysical Applications
• MoGU: Mixture-of-Gaussians with Uncertainty-based Gating for Time Series Forecasting
• Artificial Hippocampus Networks for Efficient Long-Context Modeling
• Accelerating Inference for Multilayer Neural Networks with Quantum Computers
• A Narwhal-Inspired Sensing-to-Control Framework for Small Fixed-Wing Aircraft
• MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics
• Active Control of Turbulent Airfoil Flows Using Adjoint-based Deep Learning
• Generative World Modelling for Humanoids: 1X World Model Challenge Technical Report
• HTMformer: Hybrid Time and Multivariate Transformer for Time Series Forecasting
• Machine Learning Techniques to Distinguish Giant Stars from Dwarf Stars Using Only Photometry -- Pushing Redwards
• Revealing the Temporally Stable Bimodal Energy Distribution of FRB 20121102A with a Tripled Burst Set from AI Detections
• Revisiting Mixout: An Overlooked Path to Robust Finetuning
• Multi-hop Deep Joint Source-Channel Coding with Deep Hash Distillation for Semantically Aligned Image Retrieval
• nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation
• A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT images
• Less is More: ClipBERT for Video-and-Language Learning via Sparse Sampling
• Physics-Informed Neural Network Methods for Predicting Plant Height Development
• The Boombox: Visual Reconstruction from Acoustic Vibrations
• Combining Deep Reinforcement Learning and Search for Imperfect-Information Games
• MATCH: Multi-faceted Adaptive Topo-Consistency for Semi-Supervised Histopathology Segmentation
• Automating Data-Driven Modeling and Analysis for Engineering Applications using Large Language Model Agents
• Neural Network Surrogates for Free Energy Computation of Complex Chemical Systems
• Pool Me Wisely: On the Effect of Pooling in Transformer-Based Models
• Modeling COVID-19 Dynamics in German States Using Physics-Informed Neural Networks
• How Language Models Conflate Logical Validity with Plausibility: A Representational Analysis of Content Effects
• metabeta -- A fast neural model for Bayesian mixed-effects regression
• POME: Post Optimization Model Edit via Muon-style Projection
• TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs
• Superpixel Integrated Grids for Fast Image Segmentation
• Overlap-aware segmentation for topological reconstruction of obscured objects
• msmJAX: Fast and Differentiable Electrostatics on the GPU in Python
• \bfD³QE: Learning Discrete Distribution Discrepancy-aware Quantization Error for Autoregressive-Generated Image Detection
• Stable Robot Motions on Manifolds: Learning Lyapunov-Constrained Neural Manifold ODEs
• ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamics
• Large Language Model-Based Uncertainty-Adjusted Label Extraction for Artificial Intelligence Model Development in Upper Extremity Radiography
• Permutation-Invariant Representation Learning for Robust and Privacy-Preserving Feature Selection
• RamPINN: Recovering Raman Spectra From Coherent Anti-Stokes Spectra Using Embedded Physics
• Attack of the Tails: Yes, You Really Can Backdoor Federated Learning
• Energy-Energy Flow Networks
• Boomerang Distillation Enables Zero-Shot Model Size Interpolation
• ResCP: Reservoir Conformal Prediction for Time Series Forecasting
• Diffusion²: Turning 3D Environments into Radio Frequency Heatmaps
• Demystifying Synthetic Data in LLM Pre-training: A Systematic Study of Scaling Laws, Benefits, and Pitfalls
• Vision Transformer for Transient Noise Classification
• Predictive Feature Caching for Training-free Acceleration of Molecular Geometry Generation
• From Behavioral Performance to Internal Competence: Interpreting Vision-Language Models with VLM-Lens
• GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning
• MultiModal Action Conditioned Video Generation
• TAG:Tangential Amplifying Guidance for Hallucination-Resistant Diffusion Sampling
• Differentiable physics for sound field reconstruction
• KVLinC : KV Cache Quantization with Hadamard Rotation and Linear Correction
• An End-to-End Framework for Molecular Conformation Generation via\n Bilevel Programming
• Joint Learning of Pose Regression and Denoising Diffusion with Score Scaling Sampling for Category-level 6D Pose Estimation
• A Hierarchical Self-Consistent Regularization Approach to Satellite Image Time Series Classification
• Glocal Information Bottleneck for Time Series Imputation
• Combining Motion Matching and Orientation Prediction to Animate Avatars for Consumer‐Grade VR Devices
• Learning to Predict Chaos: Curriculum-Driven Training for Robust Forecasting of Chaotic Dynamics
• PhaseFormer: From Patches to Phases for Efficient and Effective Time Series Forecasting
• Byzantine-Robust Learning on Heterogeneous Datasets via Bucketing
• Understanding Negative Samples in Instance Discriminative Self-supervised Representation Learning
• In Search of Robust Measures of Generalization
• Secure Distributed Training at Scale
• Learning Representations Through Contrastive Neural Model Checking
• Efficient Training of Spiking Neural Networks by Spike-aware Data Pruning
• Diffusion Low Rank Hybrid Reconstruction for Sparse View Medical Imaging
• Flexible and Efficient Spatio-Temporal Transformer for Sequential Visual Place Recognition
• Exact Causal Attention with 10% Fewer Operations
• RainBench: Towards Global Precipitation Forecasting from Satellite Imagery
• The Unseen Frontier: Pushing the Limits of LLM Sparsity with Surrogate-Free ADMM
• Multi-Class Support Vector Machine with Differential Privacy
• BONSAI: Structure-exploiting robust Bayesian optimization for networked black-box systems under uncertainty
• TROLL: Trust Regions improve Reinforcement Learning for Large Language Models
• Multilingual Code-Switching for Zero-Shot Cross-Lingual Intent\n Prediction and Slot Filling
• DuPLUS: Dual-Prompt Vision-Language Framework for Universal Medical Image Segmentation and Prognosis
• Enhanced Self-Distillation Framework for Efficient Spiking Neural Network Training
• HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmark
• Q-Learning with Shift-Aware Upper Confidence Bound in Non-Stationary Reinforcement Learning
• Backdoor-Powered Prompt Injection Attacks Nullify Defense Methods
• Insights into Ordinal Embedding Algorithms: A Systematic Evaluation
• Explainable but Vulnerable: Adversarial Attacks on XAI Explanation in Cybersecurity Applications
• Composite Optimization with Error Feedback: the Dual Averaging Approach
• Foundation models for equation discovery in high energy physics
• Learning Robust Diffusion Models from Imprecise Supervision
• GS-Share: Enabling High-fidelity Map Sharing with Incremental Gaussian Splatting
• Fusing Multi- and Hyperspectral Satellite Data for Harmful Algal Bloom Monitoring with Self-Supervised and Hierarchical Deep Learning
• PGMEL: Policy Gradient-based Generative Adversarial Network for Multimodal Entity Linking
• Kolmogorov-Arnold Networks in Thermoelectric Materials Design
• Estimating link level traffic emissions: enhancing MOVES with open-source data
• HyperAdaLoRA: Accelerating LoRA Rank Allocation During Training via Hypernetworks without Sacrificing Performance
• Fine-tuning LLMs with variational Bayesian last layer for high-dimensional Bayesian optimization
• An Efficient, Reliable and Observable Collective Communication Library in Large-scale GPU Training Clusters
• ElasWave: An Elastic-Native System for Scalable Hybrid-Parallel Training
• Few-shot Domain Adaptation by Causal Mechanism Transfer
• Rehearsal-free and Task-free Online Continual Learning With Contrastive Prompt
• TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting
• Enhancing Rating Prediction with Off-the-Shelf LLMs Using In-Context User Reviews
• Learning Passive Continuous-Time Dynamics with Multistep Port-Hamiltonian Gaussian Processes
• ZQBA: Zero Query Black-box Adversarial Attack
• Randomness In Neural Network Training: Characterizing The Impact of Tooling
• Efficient E(3)-equivariant framework for universal charge density prediction
• STCRpy: a software suite for T-cell receptor structure parsing, interaction profiling, and machine learning dataset preparation
• Time-series forecasting with deep learning: a survey
• Adaptive Learning of Tensor Network Structures
• R³Net:Relation-embedded Representation Reconstruction Network for\n Change Captioning
• Learning Practically Feasible Policies for Online 3D Bin Packing
• Beyond Token Probes: Hallucination Detection via Activation Tensors with ACT-ViT
• Learning Energy-based Variational Latent Prior for VAEs
• CODED-SMOOTHING: Coding Theory Helps Generalization
• Lattica: A Decentralized Cross-NAT Communication Framework for Scalable AI Inference and Training
• DA²: Depth Anything in Any Direction
• TASP: Topology-aware Sequence Parallelism
• DEPTHOR++: Robust Depth Enhancement from a Real-World Lightweight dToF and RGB Guidance
• Are neural scaling laws leading quantum chemistry astray?
• Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization
• Marginal Flow: a flexible and efficient framework for density estimation
• AiDE-Q: Synthetic Labeled Datasets Can Enhance Learning Models for Quantum Property Estimation
• A Physics-Guided Probabilistic Surrogate Modeling Framework for Digital Twins of Underwater Radiated Noise
• Iterative Residual Cross-Attention Mechanism: An Integrated Approach for Audio-Visual Navigation Tasks
• LAPIS: A Performance Portable, High Productivity Compiler Framework
• FlowMoE: A Scalable Pipeline Scheduling Framework for Distributed Mixture-of-Experts Training
• Hybrid Dual-Batch and Cyclic Progressive Learning for Efficient Distributed Training
• Noise-Guided Transport for Imitation Learning
• Per-example gradients: a new frontier for understanding and improving optimizers
• GaussianLens: Localized High-Resolution Reconstruction via On-Demand Gaussian Densification
• Building the EHR Foundation Model via Next Event Prediction
• GenVarFormer: Predicting gene expression from long-range mutations in cancer
• Parallel Heuristic Search as Inference for Actor-Critic Reinforcement Learning Models
• A Cartography of Open Collaboration in Open Source AI: Mapping Practices, Motivations, and Governance in 14 Open Large Language Model Projects
• GRACE-MoE: Grouping and Replication with Locality-Aware Routing for Efficient Distributed MoE Inference
• Intra-request branch orchestration for efficient LLM reasoning
• Meta-Learning Theory-Informed Inductive Biases using Deep Kernel Gaussian Processes
• Identifying Information-Transfer Nodes in a Recurrent Neural Network Reveals Dynamic Representations
• Fidel-TS: A High-Fidelity Benchmark for Multimodal Time Series Forecasting
• Neural Message-Passing on Attention Graphs for Hallucination Detection
• Single-Core Superscalar Optimization of Clifford Neural Layers
• BFSM: 3D Bidirectional Face-Skull Morphable Model
• Quantitative convergence of trained single layer neural networks to Gaussian processes
• CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models
• Enabling Physical AI through Biological Principles
• Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
• Hybrid Layer-Wise ANN-SNN With Surrogate Spike Encoding-Decoding Structure
• OMeGa: Joint Optimization of Explicit Meshes and Gaussian Splats for Robust Scene-Level Surface Reconstruction
• Asynchronous Policy Gradient Aggregation for Efficient Distributed Reinforcement Learning
• Experience Paper: Adopting Activity Recognition in On-demand Food Delivery Business
• Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms
• BALF: Budgeted Activation-Aware Low-Rank Factorization for Fine-Tuning-Free Model Compression
• AGNOMIN -- Architecture Agnostic Multi-Label Function Name Prediction
• Non-local Recurrent Regularization Networks for Multi-view Stereo
• Derivatives of partial eigendecomposition of a real symmetric matrix for\n degenerate cases
• ClimSat – A diffusion autoencoder model for climate-conditional satellite image editing
• Foveated Retinotopy Improves Classification and Localization in Convolutional Neural Networks
• Tasting the cake: evaluating self-supervised generalization on\n out-of-distribution multimodal MRI data
• Soft Threshold Weight Reparameterization for Learnable Sparsity
• Local Learning Matters: Rethinking Data Heterogeneity in Federated Learning
• The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes
• Neural Non-Rigid Tracking
• Contact and Human Dynamics from Monocular Video
• UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generation
• Cognitive State Inference from VR Motion via Motion Foundation Model
• An Agent-Based Framework for Automated Higher-Voice Harmony Generation
• Echo Flow Networks
• Tent: Fully Test-time Adaptation by Entropy Minimization
• Adversarial Versus Federated: An Adversarial Learning based Multi-Modality Cross-Domain Federated Medical Segmentation
• IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecasting
• Position-Blind Ptychography: Viability of image reconstruction via data-driven variational inference
• Merge Now, Regret Later: The Hidden Cost of Model Merging is Adversarial Transferability
• Multi-Scale Spatial-Temporal Hypergraph Network with Lead-Lag Structures for Stock Time Series Forecasting
• Why Alignment Must Precede Distillation: A Minimal Working Explanation
• EfficientMIL: Efficient Linear-Complexity MIL Method for WSI Classification
• AW-EL-PINNs: A Multi-Task Learning Physics-Informed Neural Network for Euler-Lagrange Systems in Optimal Control Problems
• Accuracy-Robustness Trade Off via Spiking Neural Network Gradient Sparsity Trail
• Automated extraction of fungal trophic modes from literature using BioBERT: an open pilot workflow
• FM-SIREN &amp; FM-FINER: Nyquist-Informed Frequency Multiplier for Implicit Neural Representation with Periodic Activation
• Multi-layered tensor networks for image classification
• Flow Matching for Robust Simulation-Based Inference under Model Misspecification
• HLAIIPred: cross-attention mechanism for modeling the interaction of HLA class II molecules with peptides
• Using GNN property predictors as molecule generators
• On the design space between molecular mechanics and machine learning force fields
• Score-Based Change Detection for Gradient-Based Learning Machines
• One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences
• Untangling Vascular Trees for Surgery and Interventional Radiology
• ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting
• Beyond Heuristics: Globally Optimal Configuration of Implicit Neural Representations
• RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
• PT²-LLM: Post-Training Ternarization for Large Language Models
• When Does Self-supervision Improve Few-shot Learning?
• Data-Efficient Training by Evolved Sampling
• Memory Efficient and Staleness Free Pipeline Parallel DNN Training Framework with Improved Convergence Speed
• Spatiotemporal Radar Gesture Recognition with Hybrid Spiking Neural Networks: Balancing Accuracy and Efficiency
• Learning KAN-based Implicit Neural Representations for Deformable Image Registration
• Hierarchical Representation Matching for CLIP-based Class-Incremental Learning
• Bayesian estimation of gene constraint from an evolutionary model with gene features
• Structure and transport properties of LiTFSI-based deep eutectic electrolytes from machine-learned interatomic potential simulations
• DeepRacing: Parameterized Trajectories for Autonomous Racing
• Multi-stream Convolutional Neural Network with Frequency Selection for Robust Speaker Verification
• Mining Domain Knowledge: Improved Framework towards Automatically Standardizing Anatomical Structure Nomenclature in Radiotherapy
• NxMTransformer: Semi-Structured Sparsification for Natural Language Understanding via ADMM
• Nonlinear Optimization with GPU-Accelerated Neural Network Constraints
• U-MAN: U-Net with Multi-scale Adaptive KAN Network for Medical Image Segmentation
• deepSELF: An Open Source Deep Self End-to-End Learning Framework
• HOLMES: Health OnLine Model Ensemble Serving for Deep Learning Models in Intensive Care Units
• Zero-Effort Image-to-Music Generation: An Interpretable RAG-based VLM Approach
• Neural Feature Geometry Evolves as Discrete Ricci Flow
• Context and Diversity Matter: The Emergence of In-Context Learning in World Models
• Joint graph entropy knowledge distillation for point cloud classification and robustness against corruptions
• Aurora: Towards Universal Generative Multimodal Time Series Forecasting
• Impact of Collective Behaviors of Autonomous Vehicles on Urban Traffic Dynamics: A Multi-Agent Reinforcement Learning Approach
• The AI_INFN Platform: Artificial Intelligence Development in the Cloud
• Deep learning for interval-censored failure time data from case-cohort studies
• Stable and Interpretable Jet Physics with IRC-Safe Equivariant Feature Extraction
• Teaching Transformers to Solve Combinatorial Problems through Efficient Trial &amp; Error
• Bilinear relational structure fixes reversal curse and enables consistent model editing
• Sharpness-Aware Minimization Can Hallucinate Minimizers
• Beyond Structure: Invariant Crystal Property Prediction with Pseudo-Particle Ray Diffraction
• A Unifying Framework for Parallelizing Sequential Models with Linear Dynamical Systems
• Triple-BERT: Do We Really Need MARL for Order Dispatch on Ride-Sharing Platforms?
• Dual Optimistic Ascent (PI Control) is the Augmented Lagrangian Method in Disguise
• Variational Autoencoders-based Detection of Extremes in Plant Productivity in an Earth System Model
• Compute-Optimal Quantization-Aware Training
• Wav2Arrest 2.0: Long-Horizon Cardiac Arrest Prediction with Time-to-Event Modeling, Identity-Invariance, and Pseudo-Lab Alignment
• Blockwise Hadamard high-Rank Adaptation for Parameter-Efficient LLM Fine-Tuning
• A Data-driven Typology of Vision Models from Integrated Representational Metrics
• Effective continuous equations for adaptive SGD: a stochastic analysis view
• One Model, Many Morals: Uncovering Cross-Linguistic Misalignments in Computational Moral Reasoning
• Nova: Real-Time Agentic Vision-Language Model Serving with Adaptive Cross-Stage Parallelization
• Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond
• Mammo-CLIP Dissect: A Framework for Analysing Mammography Concepts in Vision-Language Models
• Identifying Group Anchors in Real-World Group Interactions Under Label Scarcity
• Active Testing: Sample-Efficient Model Evaluation
• Deconditional Downscaling with Gaussian Processes
• Probability Distribution Collapse: A Critical Bottleneck to Compact Unsupervised Neural Grammar Induction
• Dense Semantic Matching with VGGT Prior
• RollPacker: Mitigating Long-Tail Rollouts for Fast, Synchronous RL Post-Training
• RepLLM: Toward Automatically Reproducing Network Research Results
• Sig2Model: A Boosting-Driven Model for Updatable Learned Indexes
• FlowXpert: Context-Aware Flow Embedding for Enhanced Traffic Detection in IoT Network
• Neural SDEs as Infinite-Dimensional GANs
• Inductive Predictions of Extreme Hydrologic Events in The Wabash River Watershed
• Visually Grounded Continual Learning of Compositional Phrases
• d2: Improving Reasoning in Diffusion Language Models via Trajectory Likelihood Estimation
• An Adaptor for Triggering Semi-Supervised Learning to Out-of-Box Serve Deep Image Clustering
• Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition
• The Sensitivity of Variational Bayesian Neural Network Performance to Hyperparameters
• Preparation Meets Opportunity: Enhancing Data Preprocessing for ML Training With Seneca
• A Comprehensive Evaluation of YOLO-based Deer Detection Performance on Edge Devices
• Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
• C²MIL: Synchronizing Semantic and Topological Causalities in Multiple Instance Learning for Robust and Interpretable Survival Analysis
• Attention U-Net for all-sky continuous gravitational wave searches
• One Filters All: A Generalist Filter for State Estimation
• Learning to Lead Themselves: Agentic AI in MAS using MARL
• The Syntax and Semantics of einsum
• From attribution maps to human-understandable explanations through Concept Relevance Propagation
• Open-source Stand-Alone Versatile Tensor Accelerator
• Convex Potential Flows: Universal Probability Distributions with Optimal Transport and Convex Optimization
• Frequency-domain Multi-modal Fusion for Language-guided Medical Image Segmentation
• Region-of-Interest Augmentation for Mammography Classification under Patient-Level Cross-Validation
• Convolutional Occupancy Networks
• ARD-REFSM: Enhancing Reflection Symmetry Detection with Asymmetric Denoising and Rotation Equivariance
• HyperClaim: Fine-Grained Cross-Modal Hypergraph Reasoning for Video Misinformation Detection
• On component interactions in two-stage recommender systems
• DeepViT: Towards Deeper Vision Transformer
• Conditional Convolutions for Instance Segmentation
• Multi-task Learning by Leveraging the Semantic Information
• Complementary Matrix-Gated QKAN Fast-Weight Programmers for Quantum Dynamics Forecasting
• Causal-BALD: Deep Bayesian Active Learning of Outcomes to Infer Treatment-Effects from Observational Data
• It's All Just Vectorization: einx, a Universal Notation for Tensor Operations
• Face and voice cross-modal association with learning convex feature embedding
• PrintAnything: Learning an Intermediate Representation for 3D printing G-code Generation
• Enhancing Irregular Time Series Forecasting with Continuous-Time Modeling Framework
• Physics-informed neural networks (PINNs) for fluid mechanics: A review
• FORGE: Fused On-Register Gradient Elimination for Memory-Efficient LLM Training
• PureLight: Learning Complex Luminaires with Light Tracing
• KernelGenBench: A Multi-Source and Multi-Chip Benchmark for LLM-based Kernel Generation
• Learning to Explore by Reinforcement over High-Level Options
• Flower Visitation through the Lens: Exploring the Foraging Behaviour of Bombus terrestris with a Computer Vision-Based Application
• Sparse component analysis: A method that uncovers separable computations within neural population activity
• Pushing the Envelope of Thin Crack Detection
• Spontaneous symmetry breaking and Goldstone modes for deep information propagation
• Symmetry-adapted graph neural networks for constructing molecular dynamics force fields
• MG-WFBP: Merging Gradients Wisely for Efficient Communication in Distributed Deep Learning
• Efficient Semi-Implicit Variational Inference
• GCVA: A Multiview Fusion Mechanism for Heterogeneous Data Representations
• Multi-Source Video Domain Adaptation with Temporal Attentive Moment Alignment
• Dual-side Sparse Tensor Core
• Machine Learning for Electronic Design Automation: A Survey
• Comparison of Feature Engineering and End-to-End Machine Learning for Neonatal Preictal State Classification
• Deep‐learning models of the ascending proprioceptive pathway are subject to illusions
• Prediction of motions and mooring tensions for the OC3 spar in short-crested seas using a LSTM NN model, with application to fatigue damage assessment
• Recent Weakening of the Global Radiative Feedback
• Vision Permutator: A Permutable MLP-Like Architecture for Visual Recognition
• SpheronizaTor: Spherical Voxelization for Interpretable Protein Microenvironment Modeling
• DeepKin: Predicting relatedness from low-coverage genomes and paleogenomes with convolutional neural networks
• Towards adversarial robustness with 01 loss neural networks
• GPU Memory and Utilization Estimation for Training-Aware Resource Management: Opportunities and Limitations
• Use What You Know: Causal Foundation Models with Partial Graphs
• Can LLMs capture stable human-generated sentence entropy measures?
• Archaeological Site Detection: Latest Results from a Deep Learning Based Europe Wide Hillfort Search
• Machine learning of slow collective variables and enhanced sampling via spatial techniques
• Topology Aware Neural Interpolation of Scalar Fields
• EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generation
• AraCOVID19-SSD: Arabic COVID-19 Sentiment and Sarcasm Detection Dataset
• Development of a machine learning finite-range nonlocal density functional
• Machine learning coarse-grained potentials of protein thermodynamics
• Bridging Text and Knowledge with Multi-Prototype Embedding for Few-Shot Relational Triple Extraction
• Efficient Heuristic Generation for Robot Path Planning with Recurrent Generative Model
• Extended coupled-cluster approach to twisted graphene layers
• Evaluating Synthetic Data Generation for Domain Generalization in Fetal Brain MRI Segmentation
• GINA: Neural Relational Inference From Independent Snapshots
• End-to-End Learning Local Multi-view Descriptors for 3D Point Clouds
• On new data sources for the production of official statistics
• Evaluating Logical Generalization in Graph Neural Networks
• Interference and Generalization in Temporal Difference Learning
• RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification
• Effect of Demographic Bias on Skin Lesion Classification
• Connectome-constrained networks predict neural activity across the fly visual system
• ZipStrain Enables Rapid and Precise Strain-Resolved Metagenomics
• EndoUFM: Utilizing Foundation Models for Monocular depth estimation of endoscopic images
• IntraLoss: Further Margin via Gradient-Enhancing Term for Deep Face Recognition
• Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation
• Enhancing Speech Emotion Recognition with Multi-Task Learning and Dynamic Feature Fusion
• Edge-Enhanced Vision Transformer Framework for Accurate AI-Generated Image Detection
• SlowFast Rolling-Unrolling LSTMs for Action Anticipation in Egocentric Videos
• Scaling Up Exact Neural Network Compression by ReLU Stability
• Learning Structured Representations of Entity Names using Active Learning and Weak Supervision
• 3D Shape Reconstruction from Vision and Touch
• Forecasting emergency department visits in the reference hospital of the Balearic Islands: The role of tourist and weather data
• D-LIM: A neural network for interpretable gene–gene interactions
• Advances and Open Problems in Federated Learning
• Pretrained Transformers for Text Ranking: BERT and Beyond
• A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks
• Adaptive Elastic Training for Sparse Deep Learning on Heterogeneous Multi-GPU Servers
• Towards the Development of Entropy-Based Anomaly Detection in an Astrophysics Simulation
• Optical Wavelength Guided Self-Supervised Feature Learning For Galaxy\n Cluster Richness Estimate
• Merchant Category Identification Using Credit Card Transactions
• Real‐Time MRI With Deep Learning for Efficient Evaluation of Neuromuscular Breathing Impairment
• Neural Mask Generator: Learning to Generate Adaptive Word Maskings for Language Model Adaptation
• Simulation-efficient marginal posterior estimation with swyft: stop wasting your precious time
• Towards Fast and Light-Weight Restoration of Dark Images
• Unsupervised Part Discovery via Feature Alignment
• 3D Scattering Tomography by Deep Learning with Architecture Tailored to\n Cloud Fields
• Gradient-Induced Co-Saliency Detection
• Learning to Separate Clusters of Adversarial Representations for Robust Adversarial Detection
• Positional Encoding as Spatial Inductive Bias in GANs
• Uncertainty Estimation in Deep Neural Networks for Point Cloud\n Segmentation in Factory Planning
• Biomechanical modelling of brain atrophy through deep learning
• Limitations of Normalization in Attention Mechanism
• Interpretable Early Failure Detection via Machine Learning and Trace Checking-based Monitoring
• Federated Learning with Unbiased Gradient Aggregation and Controllable Meta Updating
• Cascaded Text Generation with Markov Transformers
• Contextualized Perturbation for Textual Adversarial Attack
• Text-guided Legal Knowledge Graph Reasoning
• Differentiating a Tensor Language
• ISACL: Internal State Analyzer for Copyrighted Training Data Leakage
• Transformer Modeling for Both Scalability and Performance in Multivariate Time Series
• CompLLM: Compression for Long Context Q&A
• MsFIN: Multi-scale Feature Interaction Network for Traffic Accident Anticipation
• Bridging Computational Social Science and Deep Learning: Cultural Dissemination-Inspired Graph Neural Networks
• Improving Credit Card Fraud Detection through Transformer-Enhanced GAN Oversampling
• How Machine Learning Predicts Fluid Densities under Nanoconfinement
• Choose a Transformer: Fourier or Galerkin
• Integrating Stacked Intelligent Metasurfaces and Power Control for Dynamic Edge Inference via Over-The-Air Neural Networks
• ViG-LRGC: Vision Graph Neural Networks with Learnable Reparameterized Graph Construction
• TimeMosaic: Temporal Heterogeneity Guided Time Series Forecasting via Adaptive Granularity Patch and Segment-wise Decoding
• TsqLoRA: Towards Sensitivity and Quality Low-Rank Adaptation for Efficient Fine-Tuning
• Differentiable Light Transport with Gaussian Surfels via Adapted Radiosity for Efficient Relighting and Geometry Reconstruction
• Machine learning approach to single-shot multiparameter estimation for the non-linear Schrödinger equation
• BatchTNMC: Efficient sampling of two-dimensional spin glasses using tensor network Monte Carlo
• Towards Practical Multi-label Causal Discovery in High-Dimensional Event Sequences via One-Shot Graph Aggregation
• On the model-based stochastic value gradient for continuous reinforcement learning
• MusPy: A Toolkit for Symbolic Music Generation
• WolBanking77: Wolof Banking Speech Intent Classification Dataset
• PPG-Distill: Efficient Photoplethysmography Signals Analysis via Foundation Model Distillation
• VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction
• Zero-Shot Transferable Solution Method for Parametric Optimal Control Problems
• All-magnonic neurons for analog artificial neural networks
• Learning to Condition: A Neural Heuristic for Scalable MPE Inference
• TensLoRA: Tensor Alternatives for Low-Rank Adaptation
• Automated Labeling of Intracranial Arteries with Uncertainty Quantification Using Deep Learning
• SeqBattNet: A Discrete-State Physics-Informed Neural Network with Aging Adaptation for Battery Modeling
• ControlEchoSynth: Boosting Ejection Fraction Estimation Models via Controlled Video Diffusion
• MetaEmbed: Scaling Multimodal Retrieval at Test-Time with Flexible Late Interaction
• CSDformer: A Conversion Method for Fully Spike-Driven Transformer
• SC2Tools: StarCraft II Toolset and Dataset API
• Conv-like Scale-Fusion Time Series Transformer: A Multi-Scale Representation for Variable-Length Long Time Series
• LoT-Pass: Long-term-robust Image Watermarking for Image to Video Generation
• Probabilistic Token Alignment for Large Language Model Fusion
• What's a good imputation to predict with missing values?
• MTS-DMAE: Dual-Masked Autoencoder for Unsupervised Multivariate Time Series Representation Learning
• PTQTP: Post-Training Quantization to Trit-Planes for Large Language Models
• Scaling Law for Recommendation Models: Towards General-purpose User Representations
• Self-Supervised Discovery of Neural Circuits in Spatially Patterned Neural Responses with Graph Neural Networks
• Automatic Classification of Magnetic Chirality of Solar Filaments from H-Alpha Observations
• Looking in the mirror: A faithful counterfactual explanation method for interpreting deep image classification models
• Mixture of Noise for Pre-Trained Model-Based Class-Incremental Learning
• DoubleGen: Debiased Generative Modeling of Counterfactuals
• CGTGait: Collaborative Graph and Transformer for Gait Emotion Recognition
• Pico: A Modular Framework for Hypothesis-Driven Small Language Model Research
• "It Was a Magical Box": Understanding Practitioner Workflows and Needs in Optimization
• CoUn: Empowering Machine Unlearning via Contrastive Learning
• Improving Deep Tabular Learning
• CausalSent: Interpretable Sentiment Classification with RieszNet
• Towards Optimal Convolutional Transfer Learning Architectures for Breast Lesion Classification and ACL Tear Detection
• TOFLUX: A Differentiable Topology Optimization Framework for Multiphysics Fluidic Problems
• Sequential Token Merging: Revisiting Hidden States
• Inverse Optimization Latent Variable Models for Learning Costs Applied to Route Problems
• Learning Graph Models for Retrosynthesis Prediction
• Targeted Fine-Tuning of DNN-Based Receivers via Influence Functions
• Reverse Engineering of Music Mixing Graphs with Differentiable Processors and Iterative Pruning
• Differentiable Acoustic Radiance Transfer
• A Flow-rate-conserving CNN-based Domain Decomposition Method for Blood Flow Simulations
• ToFU: Transforming How Federated Learning Systems Forget User Data
• GPU Temperature Simulation-Based Testing for In-Vehicle Deep Learning Frameworks
• GS-Scale: Unlocking Large-Scale 3D Gaussian Splatting Training via Host Offloading
• De-crackling Virtual Analog Controls with Asymptotically Stable Recurrent Neural Networks
• SuperGen: An Efficient Ultra-high-resolution Video Generation System with Sketching and Tiling
• Beyond Words: Enhancing Desire, Emotion, and Sentiment Recognition with Non-Verbal Cues
• MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
• Distributed Multi-Task Learning for Joint Wireless Signal Enhancement and Recognition
• Accelerating Atomic Fine Structure Determination with Graph Reinforcement Learning
• AdaSports-Traj: Role- and Domain-Aware Adaptation for Multi-Agent Trajectory Modeling in Sports
• Distribution Estimation for Global Data Association via Approximate Bayesian Inference
• Robust deep learning–based protein sequence design using ProteinMPNN
• Lightweight and Accurate Multi-View Stereo with Confidence-Aware Diffusion Model
• Multi-Task Learning of Query Intent and Named Entities using Transfer\n Learning
• Balancing Sparse RNNs with Hyperparameterization Benefiting Meta-Learning
• Differentiating Through a Quadratic Cone Program
• Learning Interpretable Differentiable Logic Networks for Time-Series Classification
• Hint: hierarchical inter-frame correlation for one-shot point cloud sequence compression
• SINAI at eRisk@CLEF 2022: Approaching Early Detection of Gambling and Eating Disorders with Natural Language Processing
• FlowCast-ODE: Continuous Hourly Weather Forecasting with Dynamic Flow Matching and ODE Solver
• Pre-training Autoencoder for Acoustic Event Classification via Blinky
• Neural Earthquake Forecasting with Minimal Information: Limits, Interpretability, and the Role of Markov Structure
• Evaluating the Effectiveness of Coverage-Guided Fuzzing for Testing Deep Learning Library APIs
• No-Press Diplomacy from Scratch
• Super-Linear: A Lightweight Pretrained Mixture of Linear Experts for Time Series Forecasting
• OpenViGA: Video Generation for Automotive Driving Scenes by Streamlining and Fine-Tuning Open Source Models with Public Data
• Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentation
• Causal Reasoning Elicits Controllable 3D Scene Generation
• Disentangled Geometry and Appearance for Efficient Multi-View Surface Reconstruction and Rendering
• Fair-GPTQ: Bias-Aware Quantization for Large Language Models
• Adversarial Distilled Retrieval-Augmented Guarding Model for Online Malicious Intent Detection
• Neutral Face Game Character Auto-Creation via PokerFace-GAN
• Progressive Identification of True Labels for Partial-Label Learning
• Generative AI for Misalignment-Resistant Virtual Staining to Accelerate Histopathology Workflows
• Exploring the Relationship between Brain Hemisphere States and Frequency Bands through Deep Learning Optimization Techniques
• AIN: Fast and Accurate Sequence Labeling with Approximate Inference Network
• Kernel methods through the roof: handling billions of points efficiently
• VocSegMRI: Multimodal Learning for Precise Vocal Tract Segmentation in Real-time MRI
• Semantic Graph Based Place Recognition for 3D Point Clouds
• Unsupervised Discovery of 3D Physical Objects from Video
• A Regression Testing Framework with Automated Assertion Generation for Machine Learning Notebooks
• Bridging Past and Future: Distribution-Aware Alignment for Time Series Forecasting
• GraphMend: Code Transformations for Fixing Graph Breaks in PyTorch 2
• Learning Progression-Guided AI Evaluation of Scientific Models To Support Diverse Multi-Modal Understanding in NGSS Classroom
• Federated CycleGAN for Privacy-Preserving Image-to-Image Translation
• Improving 3D Gaussian Splatting Compression by Scene-Adaptive Lattice Vector Quantization
• Quickly Tuning Foundation Models for Image Segmentation
• Density-Aware Farthest Point Sampling
• Weakly and Self-Supervised Class-Agnostic Motion Prediction for Autonomous Driving
• Towards Open World Object Detection
• Deep Learning for Analyzing Chaotic Dynamics in Biological Time Series: Insights from Frog Heart Signals
• Spatiotemporal graph neural process for reconstruction, extrapolation, and classification of cardiac trajectories
• CoFiNet: Reliable Coarse-to-fine Correspondences for Robust Point Cloud Registration
• BackPACK: Packing more into backprop
• WeNet: Production oriented Streaming and Non-streaming End-to-End Speech Recognition Toolkit
• In-Place Scene Labelling and Understanding with Implicit Scene Representation
• Proposal Learning for Semi-Supervised Object Detection
• Path Integral Sampler: a stochastic control approach for sampling
• Learning to Control PDEs with Differentiable Physics
• NORESQA: A Framework for Speech Quality Assessment using Non-Matching References
• Designing Rules to Pick a Rule: Aggregation by Consistency
• How BPE Affects Memorization in Transformers
• Intrinsic-Extrinsic Preserved GANs for Unsupervised 3D Pose Transfer
• Few-NERD: A Few-Shot Named Entity Recognition Dataset
• PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
• Camouflaged Object Segmentation with Distraction Mining
• Unsupervised Domain Adaptation of Black-Box Source Models
• Uncovering and Mitigating Destructive Multi-Embedding Attacks in Deepfake Proactive Forensics
• MeLT: Message-Level Transformer with Masked Document Representations as\n Pre-Training for Stance Detection
• C5T5: Controllable Generation of Organic Molecules with Transformers
• Text-Based Person Search with Limited Data
• Learning to Extend Molecular Scaffolds with Structural Motifs
• Few-Shot Learning with Intra-Class Knowledge Transfer
• ConE: Cone Embeddings for Multi-Hop Reasoning over Knowledge Graphs
• What Robot do I Need? Fast Co-Adaptation of Morphology and Control using Graph Neural Networks
• Graph Neural Networks Including Sparse Interpretability
• SamudrACE: Fast and Accurate Coupled Climate Modeling with 3D Ocean and Atmosphere Emulators
• Comparative Analysis of Wave Scattering Numerical Modeling Using the Boundary Element Method and Physics-Informed Neural Networks
• Dynamic Relational Priming Improves Transformer in Multivariate Time Series
• RealSmileNet: A Deep End-To-End Network for Spontaneous and Posed Smile Recognition
• Artificial Neural Networks for Neuroscientists: A Primer
• Is Attention Better Than Matrix Decomposition?
• Robust Fetal Pose Estimation across Gestational Ages via Cross-Population Augmentation
• Jet tagging in the Lund plane with graph networks
• SpeechBrain: A General-Purpose Speech Toolkit
• Detecting Multilevel Manipulation from Limit Order Book via Cascaded Contrastive Representation Learning
• PD-Loss: Proxy-Decidability for Efficient Metric Learning
• MSMA: Multi-Scale Feature Fusion For Multi-Attribute 3D Face Reconstruction From Unconstrained Images
• Compositional Generalization by Learning Analytical Expressions
• Deep UAV Localization with Reference View Rendering
• Efficient Byzantine-Robust Privacy-Preserving Federated Learning via Dimension Compression
• Reconstructing High-fidelity Plasma Turbulence with Data-driven Tuning of Diffusion in Low Resolution Grids
• ESPnet2-TTS: Extending the Edge of TTS Research
• Momentum Contrastive Autoencoder: Using Contrastive Learning for Latent Space Distribution Matching in WAE
• Optimizing Quantum Photonic Integrated Circuits using Differentiable Tensor Networks
• From Presence‐Only to Abundance Species Distribution Models Using Transfer Learning
• Neural networks in the search for fast radio bursts with RATAN-600
• Your Compiler is Backdooring Your Model: Understanding and Exploiting Compilation Inconsistency Vulnerabilities in Deep Learning Compilers
• Chameleon: Taming Dynamic Operator Sequences for Memory-Intensive LLM Training
• LibEMER: A novel benchmark and algorithms library for EEG-based Multimodal Emotion Recognition
• Investigating the Lottery Ticket Hypothesis for Variational Quantum Circuits
• Real-time reinforcement learning for turbulent state-dependent control in a bluff-body wake
• COIN: COmpression with Implicit Neural representations
• Towards Automated Error Discovery: A Study in Conversational AI
• GoldenTransformer: A Modular Fault Injection Framework for Transformer Robustness Research
• Language-based Color ISP Tuning
• RSL-RL: A Learning Library for Robotics Research
• QVHighlights: Detecting Moments and Highlights in Videos via Natural Language Queries
• Sound Matching an Analogue Levelling Amplifier Using the Newton-Raphson Method
• Context Copying Modulation: The Role of Entropy Neurons in Managing Parametric and Contextual Knowledge Conflicts
• pySigLib -- Fast Signature-Based Computations on CPU and GPU
• AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis
• Developer-LLM Conversations: An Empirical Study of Interactions and Generated Code Quality
• Physics-informed sensor coverage through structure preserving machine learning
• PatchCleanser: Certifiably Robust Defense against Adversarial Patches for Any Image Classifier
• DiffAero: A GPU-Accelerated Differentiable Simulation Framework for Efficient Quadrotor Policy Learning
• A Certifiable Machine Learning-Based Pipeline to Predict Fatigue Life of Aircraft Structures
• Loss Behavior in Supervised Learning with Entangled States
• Semi-Automating Knowledge Base Construction for Cancer Genetics
• Symbolic Feedforward Networks for Probabilistic Finite Automata: Exact Simulation and Learnability
• Neural Scaling Laws for Deep Regression
• Controllable Neural Dialogue Summarization with Personal Named Entity Planning
• Quantum parameter estimation with uncertainty quantification from continuous measurement data using neural network ensembles
• Service Function Chaining Architecture for Multi-hop Split Inference and Learning
• DOSA: Differentiable Model-Based One-Loop Search for DNN Accelerators
• Exploring Expert Specialization through Unsupervised Training in Sparse Mixture of Experts
• ButterflyQuant: Ultra-low-bit LLM Quantization through Learnable Orthogonal Butterfly Transforms
• Geometric Neural Distance Fields for Learning Human Motion Priors
• Identifying Mislabeled Data using the Area Under the Margin Ranking
• NAT: Learning to Attack Neurons for Enhanced Adversarial Transferability
• A neural drift-plus-penalty algorithm for network power allocation and routing
• Boosting Data Utilization for Multilingual Dense Retrieval
• Channel Estimation and Analog Precoding for Pixel-based Fluid-Antenna-Assisted Multiuser MIMO-OFDM Systems
• Modular, On-Site Solutions with Lightweight Anomaly Detection for Sustainable Nutrient Management in Agriculture
• Sensitivity-LoRA: Low-Load Sensitivity-Based Fine-Tuning for Large Language Models
• MDIQA: Unified Image Quality Assessment for Multi-dimensional Evaluation and Restoration
• Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction
• Images in Motion?: A First Look into Video Leakage in Collaborative Deep Learning
• CoSwin: Convolution Enhanced Hierarchical Shifted Window Attention For Small-Scale Vision
• TensorFI: A Flexible Fault Injection Framework for TensorFlow Applications
• SAFT: Shape and Appearance of Fabrics from Template via Differentiable Physical Simulations from Monocular Video
• Interior Point Solving for LP-based prediction+optimisation
• Model-Agnostic Learning to Meta-Learn
• Learning Turbulent Flows with Generative Models: Super-resolution, Forecasting, and Sparse Flow Reconstruction
• Compressing CNN models for resource-constrained systems by channel and layer pruning
• FractalPINN-Flow: A Fractal-Inspired Network for Unsupervised Optical Flow Estimation with Total Variation Regularization
• Towards Open-Vocabulary Multimodal 3D Object Detection with Attributes
• Augmenting Neural Networks-Based Model Approximators in Robotic Force-Tracking Tasks
• S²Transformer: Scalable Structured Transformers for Global Station Weather Forecasting
• Value bounds and Convergence Analysis for Averages of LRP attributions
• Reinforcement-Guided Hyper-Heuristic Hyperparameter Optimization for Fair and Explainable Spiking Neural Network-Based Financial Fraud Detection
• Towards Communication-Efficient Decentralized Federated Graph Learning over Non-IID Data
• Prototypical Representation Learning for Relation Extraction
• ArtifactGen: Benchmarking WGAN-GP vs Diffusion for Label-Aware EEG Artifact Synthesis
• SVN-ICP: Uncertainty Estimation of ICP-based LiDAR Odometry using Stein Variational Newton
• Multimodal Contrastive Pretraining of CBCT and IOS for Enhanced Tooth Segmentation
• ScoreHOI: Physically Plausible Reconstruction of Human-Object Interaction via Score-Guided Diffusion
• Algorithmic differentiation for plane-wave DFT: materials design, error control and learning model parameters
• Enhancing Fiber Orientation Distributions using convolutional Neural\n Networks
• Variational Depth Search in ResNets
• SEEC: Segmentation-Assisted Multi-Entropy Models for Learned Lossless Image Compression
• A Generalisable Generative Model for Multi-Detector Calorimeter Simulation
• Hoplite: Efficient and Fault-Tolerant Collective Communication for Task-Based Distributed Systems
• Basis Vector Metric: A Method for Robust Open-Ended State Change Detection
• XSRD-Net: EXplainable Stroke Relapse Detection
• AdaMixT: Adaptive Weighted Mixture of Multi-Scale Expert Transformers for Time Series Forecasting
• RINO: Renormalization Group Invariance with No Labels
• Astra: A Multi-Agent System for GPU Kernel Performance Optimization
• Long Document Ranking with Query-Directed Sparse Transformer
• DKM: Differentiable K-Means Clustering Layer for Neural Network Compression
• Sparse Uncertainty Representation in Deep Learning with Inducing Weights
• Neural Cone Radiosity for Interactive Global Illumination with Glossy Materials
• Strength from Weakness: Fast Learning Using Weak Supervision
• Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition
• Towards Robust Graph Contrastive Learning
• Stochastic Aggregation in Graph Neural Networks
• Coordinate Attention for Efficient Mobile Network Design
• Learning Generalized Hamiltonian Dynamics with Stability from Noisy Trajectory Data
• IP-Basis PINNs: Efficient Multi-Query Inverse Parameter Estimation
• PLaID++: A Preference Aligned Language Model for Targeted Inorganic Materials Design
• Learning Long-Term Dependencies in Irregularly-Sampled Time Series
• A machine-learned expression for the excess Gibbs energy
• MS2: Multi-Document Summarization of Medical Studies
• Guiding Cascading Failure Search with Interpretable Graph Convolutional Network
• Deep Learning for Markov Chains: Lyapunov Functions, Poisson's Equation, and Stationary Distributions
• NeuroDeX: Unlocking Diverse Support in Decompiling Deep Neural Network Executables
• Unsupervised Domain Adaptation in the Absence of Source Data
• Equipping SBMs with RBMs: An Explainable Approach for Analysis of Networks with Covariates
• Back To The Drawing Board: Rethinking Scene-Level Sketch-Based Image Retrieval
• Investigating Location-Regularised Self-Supervised Feature Learning for Seafloor Visual Imagery
• Robust and Adaptive Spectral Method for Representation Multi-Task Learning with Contamination
• An Interpretable AI Framework to Disentangle Self-Interacting and Cold Dark Matter in Galaxy Clusters: The CKAN Approach
• Benchmarking Information Retrieval Models on Complex Retrieval Tasks
• Albumentations: Fast and Flexible Image Augmentations
• Self-Adaptive Training: beyond Empirical Risk Minimization
• Micro-Expression Recognition via Fine-Grained Dynamic Perception
• Modeling Magnetoelastic Wave Interactions in Magnetic Films and Heterostructures: A finite-difference approach
• Going Beyond Linear Transformers with Recurrent Fast Weight Programmers
• Pyserini: An Easy-to-Use Python Toolkit to Support Replicable IR Research with Sparse and Dense Representations
• Physics-Guided Diffusion Transformer with Spherical Harmonic Posterior Sampling for High-Fidelity Angular Super-Resolution in Diffusion MRI
• A guide to machine learning for biologists
• WDNet: Watermark-Decomposition Network for Visible Watermark Removal
• Discovering Symbolic Models from Deep Learning with Inductive Biases
• Ekya: Continuous Learning of Video Analytics Models on Edge Compute\n Servers
• MonoGlass3D: Monocular 3D Glass Detection with Plane Regression and Adaptive Feature Fusion
• Viewmaker Networks: Learning Views for Unsupervised Representation Learning
• Bridging the Gap Between Theoretical and Practical Reinforcement Learning in Undergraduate Education
• Depth-Aware Super-Resolution via Distance-Adaptive Variational Formulation
• BLK-REW: A Unified Block-based DNN Pruning Framework using Reweighted Regularization Method
• Causal Multi-fidelity Surrogate Forward and Inverse Models for ICF Implosions
• Robust Model Predictive Control Design for Autonomous Vehicles with Perception-based Observers
• LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood
• Efficient iPEPS Simulation on the Honeycomb Lattice via QR-based CTMRG
• Systematic Review and Meta-analysis of AI-driven MRI Motion Artifact Detection and Correction
• Space-time Mixing Attention for Video Transformer
• Deep-RLS: A Model-Inspired Deep Learning Approach to Nonlinear PCA
• Neuro-Spectral Architectures for Causal Physics-Informed Networks
• Linear Mode Connectivity in Multitask and Continual Learning
• Elucidating the Design Space of Decay in Linear Attention
• Kinetics of Barrier Crossing Events from Temperature Accelerated Sliced Sampling Simulations
• veScale: Consistent and Efficient Tensor Programming with Eager-Mode SPMD
• Data Poisoning Attacks Against Federated Learning Systems
• Semantic Bilinear Pooling for Fine-Grained Recognition
• VCMamba: Bridging Convolutions with Multi-Directional Mamba for Efficient Visual Representation
• Measuring the Measures: Discriminative Capacity of Representational Similarity Metrics Across Model Families
• Inpaint4Drag: Repurposing Inpainting Models for Drag-Based Image Editing via Bidirectional Warping
• Generative Models as Distributions of Functions
• Rethinking PointNet Embedding for Faster and Compact Model
• Learning neural representations for X-ray ptychography reconstruction with unknown probes
• An analog-electronic implementation of a harmonic oscillator recurrent neural network
• Disentangled Multi-modal Learning of Histology and Transcriptomics for Cancer Characterization
• Learning from Majority Label: A Novel Problem in Multi-class Multiple-Instance Learning
• LATTE: A Decoding Architecture for Quantum Computing with Temporal and Spatial Scalability
• Deep Proxy Causal Learning and its Application to Confounded Bandit\n Policy Evaluation
• ResiliNet: Failure-Resilient Inference in Distributed Neural Networks
• Anti-establishment sentiment on TikTok: Implications for understanding influence(rs) and expertise on social media
• TyXe: Pyro-based Bayesian neural nets for Pytorch
• Optimizing Frequent Checkpointing via Low-Cost Differential for Distributed Training Systems
• Tongji University Team for the VoxCeleb Speaker Recognition Challenge 2020
• Tracking by Joint Local and Global Search: A Target-aware Attention based Approach
• Who Should Go First? A Self-Supervised Concept Sorting Model for Improving Taxonomy Expansion
• Learning Convex Optimization Control Policies
• The Optimiser Hidden in Plain Sight: Training with the Loss Landscape's Induced Metric
• Searching for HWW Anomalous Couplings with Simulation-Based Inference
• Uncertainty-driven Adaptive Exploration
• LSAM: Asynchronous Distributed Training with Landscape-Smoothed Sharpness-Aware Minimization
• LogME: Practical Assessment of Pre-trained Models for Transfer Learning
• Neural marching cubes
• TRADI: Tracking deep neural network weight distributions for uncertainty\n estimation
• Congestion-aware Multi-agent Trajectory Prediction for Collision Avoidance
• Gradient Estimation Methods of Approximate Multipliers for High-Accuracy Retraining of Deep Learning Models
• Parallel-Constraint Model Predictive Control: Exploiting Parallel Computation for Improving Safety
• Stochastic versus Deterministic in Stochastic Gradient Descent
• Positive-Congruent Training: Towards Regression-Free Model Updates
• Anchor-Free Person Search
• Beyond Self-attention: External Attention using Two Linear Layers for Visual Tasks
• The EPIC-KITCHENS Dataset: Collection, Challenges and Baselines
• Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection
• Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery
• Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins
• SceneGen: Learning to Generate Realistic Traffic Scenes
• On the Generalization of Models Trained with SGD: Information-Theoretic Bounds and Implications
• Robots of the Lost Arc: Self-Supervised Learning to Dynamically Manipulate Fixed-Endpoint Cables
• Pipeline Parallelism for Inference on Heterogeneous Edge Computing
• SIM-ECG: A Signal Importance Mask-driven ECGClassification System
• Deep Diacritization: Efficient Hierarchical Recurrence for Improved Arabic Diacritization
• Self-Validated Learning for Particle Separation: A Correctness-Based Self-Training Framework Without Human Labels
• Frame Averaging for Invariant and Equivariant Network Design
• Improving atomic force microscopy structure discovery via style-translation
• Deep learning-enabled virtual multiplexed immunostaining of label-free tissue for vascular invasion assessment
• Training GANs with Stronger Augmentations via Contrastive Discriminator
• Causal Discovery in Physical Systems from Videos
• Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularization
• Differentiable Expectation-Maximisation and Applications to Gaussian Mixture Model Optimal Transport
• Generalized Zero-Shot Domain Adaptation via Coupled Conditional Variational Autoencoders
• Dual-Resolution Correspondence Networks
• Comparing SNNs and RNNs on Neuromorphic Vision Datasets: Similarities and Differences
• Spatial-Adaptive Network for Single Image Denoising
• FuXi-TC: A generative framework integrating deep learning and physics-based models for improved tropical cyclone forecasts
• Towards Full-line Code Completion with Neural Language Models
• A Runtime-Based Computational Performance Predictor for Deep Neural Network Training
• ADVMEM: Adversarial Memory Initialization for Realistic Test-Time Adaptation via Tracklet-Based Benchmarking
• GradES: Significantly Faster Training in Transformers with Gradient-Based Early Stopping
• HodgeFormer: Transformers for Learnable Operators on Triangular Meshes through Data-Driven Hodge Matrices
• GRNet: Gridding Residual Network for Dense Point Cloud Completion
• Adversarial Stylometry in the Wild: Transferable Lexical Substitution\n Attacks on Author Profiling
• Fail2Progress: Learning from Real-World Robot Failures with Stein Variational Inference
• pixelNeRF: Neural Radiance Fields from One or Few Images
• Conformal Predictive Monitoring for Multi-Modal Scenarios
• Comparison between Supervised and Unsupervised Learning in Deep Unfolded Sparse Signal Recovery
• Enhanced Universal Dependency Parsing with Automated Concatenation of Embeddings
• REFINESTAT: Efficient Exploration for Probabilistic Program Synthesis
• Leveraging learned representations and multitask learning for lysine methylation site discovery
• Sketch-Driven Regular Expression Generation from Natural Language and Examples
• Street-Level Geolocalization Using Multimodal Large Language Models and Retrieval-Augmented Generation
• MD-PNOP: Equation-Recast Neural Operators for Minimal-Data Extrapolation and PDE Solver Acceleration
• Summarize and Search: Learning Consensus-aware Dynamic Convolution for Co-Saliency Detection
• DiffCoord: Differentiable Coordination for Distributed Multi-Agent Trajectory Optimization
• Trait specialization facilitates autonomous selfing ability in a mixed‐mating plant
• SECANT: Self-Expert Cloning for Zero-Shot Generalization of Visual Policies
• Self-Supervised Learning for Gastritis Detection with Gastric X-ray Images
• Deep Ensemble Collaborative Learning by using Knowledge-transfer Graph for Fine-grained Object Classification
• Wavelet-Enhanced PaDiM for Industrial Anomaly Detection
• What Expressivity Theory Misses: Message Passing Complexity for GNNs
• Learning Slice-Aware Representations with Mixture of Attentions
• CascadeFormer: A Family of Two-stage Cascading Transformers for Skeleton-based Human Action Recognition
• HADIS: Hybrid Adaptive Diffusion Model Serving for Efficient Text-to-Image Generation
• Automatic airway segmentation from Computed Tomography using robust and efficient 3-D convolutional neural networks
• AI-driven Dispensing of Coral Reseeding Devices for Broad-scale Restoration of the Great Barrier Reef
• SegDINO: An Efficient Design for Medical and Natural Image Segmentation with DINO-V3
• Pose as Clinical Prior: Learning Dual Representations for Scoliosis Screening
• Partial success in closing the gap between human and machine vision
• We are More than Our Joints: Predicting how 3D Bodies Move
• Disentangling Slow and Fast Temporal Dynamics in Degradation Inference with Hierarchical Differential Models
• SmartFLow: A Communication-Efficient SDN Framework for Cross-Silo Federated Learning
• Application of discrete Ricci curvature in pruning randomly wired neural networks: A case study with chest x-ray classification of COVID-19
• Reinforcement Learning of Dolly-In Filming Using a Ground-Based Robot
• A Deep Learning Framework for Joint Channel Acquisition and Communication Optimization in Movable Antenna Systems
• Color2Embed: Fast Exemplar-Based Image Colorization using Color Embeddings
• SABR: A Stable Adaptive Bitrate Framework Using Behavior Cloning Pretraining and Reinforcement Learning Fine-Tuning
• LUT-Fuse: Towards Extremely Fast Infrared and Visible Image Fusion via Distillation to Learnable Look-Up Tables
• TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models
• Gradient-based Hyperparameter Optimization Over Long Horizons
• Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse\n Coding
• Learning to Shard: RL for Co-optimizing the Parallelism Degrees and Per-operator Sharding Dimensions in Distributed LLM Inference
• First Order Model-Based RL through Decoupled Backpropagation
• CNN with large memory layers
• Unsupervised Video Continual Learning via Non-Parametric Deep Embedded Clustering
• Activation Subspaces for Out-of-Distribution Detection
• On the Hardness of Learning GNN-based SAT Solvers: The Role of Graph Ricci Curvature
• SIMILAR: Submodular Information Measures Based Active Learning In\n Realistic Scenarios
• Independent Prototype Propagation for Zero-Shot Compositionality
• Federated Multi-Task Learning under a Mixture of Distributions
• Towards On-Device Personalization: Cloud-device Collaborative Data Augmentation for Efficient On-device Language Model
• PDTrim: Targeted Pruning for Prefill-Decode Disaggregation in Inference
• Scalable Equilibrium Propagation via Intermediate Error Signals for Deep Convolutional CRNNs
• Automated Multi-label Classification of Eleven Retinal Diseases: A Benchmark of Modern Architectures and a Meta-Ensemble on a Large Synthetic Dataset
• Multimodal Fusion Refiner Networks
• EXPATS: A Toolkit for Explainable Automated Text Scoring
• Learning to Assemble the Soma Cube with Legal-Action Masked DQN and Safe ZYZ Regrasp on a Doosan M0609
• Accelerating Mixture-of-Experts Inference by Hiding Offloading Latency with Speculative Decoding
• Batch Normalization Provably Avoids Rank Collapse for Randomly Initialised Deep Networks
• Differentiable Weighted Finite-State Transducers
• Mind the Pad -- CNNs can Develop Blind Spots
• PAIR: Planning and Iterative Refinement in Pre-trained Transformers for Long Text Generation
• Hybrid Monte Carlo Metadynamics (hybridMC-MetaD)
• Efficient Semi-Supervised Gross Target Volume of Nasopharyngeal Carcinoma Segmentation via Uncertainty Rectified Pyramid Consistency
• Estimating Model Uncertainty of Neural Networks in Sparse Information\n Form
• Collective Learning by Ensembles of Altruistic Diversifying Neural Networks
• ViT-V-Net: Vision Transformer for Unsupervised Volumetric Medical Image Registration
• ChainReaction: Causal Chain-Guided Reasoning for Modular and Explainable Causal-Why Video Question Answering
• Cosmo-Learn: code for learning cosmology using different methods and mock data
• Meta-learning ecological priors from large language models explains human learning and decision making
• ATM-GAD: Adaptive Temporal Motif Graph Anomaly Detection for Financial Transaction Networks
• CIGaRS I: Combined simulation-based inference from SNae Ia and host photometry
• Automated simulation-based design via multi-fidelity active learning and optimisation for laser direct drive implosions
• SKGE-SWIN: End-To-End Autonomous Vehicle Waypoint Prediction and Navigation Using Skip Stage Swin Transformer
• CraftGraffiti: Exploring Human Identity with Custom Graffiti Art via Facial-Preserving Diffusion Models
• Embracing Aleatoric Uncertainty: Generating Diverse 3D Human Motion
• Density Deconvolution with Normalizing Flows
• Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts
• PFA: Privacy-preserving Federated Adaptation for Effective Model Personalization
• Machine-Learning-Assisted Pulse Design for State Preparation in a Noisy Environment
• Visualizing Color-wise Saliency of Black-Box Image Classification Models
• Blind Image Super-resolution with Elaborate Degradation Modeling on Noise and Kernel
• Metatensor and metatomic: foundational libraries for interoperable atomistic machine learning
• NiceWebRL: a Python library for human subject experiments with reinforcement learning environments
• Seam360GS: Seamless 360° Gaussian Splatting from Real-World Omnidirectional Images
• Weisfeiler and Lehman Go Cellular: CW Networks
• Financial Decision Making using Reinforcement Learning with Dirichlet Priors and Quantum-Inspired Genetic Optimization
• GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity
• ASCMamba: Multimodal Time-Frequency Mamba for Acoustic Scene Classification
• Finding and Fixing Spurious Patterns with Explanations
• Constructing Geographic and Long-term Temporal Graph for Traffic Forecasting
• Symplectic convolutional neural networks
• WaveHiT-SR: Hierarchical Wavelet Network for Efficient Image Super-Resolution
• Physics-Informed DeepONet Coupled with FEM for Convective Transport in Porous Media with Sharp Gaussian Sources
• Scalable Rule-Based Representation Learning for Interpretable Classification
• Generalized Planning With Deep Reinforcement Learning
• FreeVPS: Repurposing Training-Free SAM2 for Generalizable Video Polyp Segmentation
• ALSA: Anchors in Logit Space for Out-of-Distribution Accuracy Estimation
• Understanding and Mitigating Exploding Inverses in Invertible Neural\n Networks
• Cross-Domain Few-Shot Learning by Representation Fusion
• Optimizing Functionals on the Space of Probabilities with Input Convex\n Neural Networks
• Learning from Similarity-Confidence Data
• Diffusion Normalizing Flow
• Amplifying Emotional Signals: Data-Efficient Deep Learning for Robust Speech Emotion Recognition
• Differentiating through Log-Log Convex Programs
• Random forest-based out-of-distribution detection for robust lung cancer segmentation
• TEASEL: A Transformer-Based Speech-Prefixed Language Model
• Solon: Communication-efficient Byzantine-resilient Distributed Training via Redundant Gradients
• Breaking the Black Box: Inherently Interpretable Physics-Constrained Machine Learning With Weighted Mixed-Effects for Imbalanced Seismic Data
• On the Generalisation of Koopman Representations for Chaotic System Control
• Distance-informed Neural Processes
• pyFAST: A Modular PyTorch Framework for Time Series Modeling with Multi-source and Sparse Data
• Cross-Domain Gradient Discrepancy Minimization for Unsupervised Domain Adaptation
• ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive
• SEA: Sentence Encoder Assembly for Video Retrieval by Textual Queries
• Multi-task Reinforcement Learning with a Planning Quasi-Metric
• Assessment of Reward Functions in Reinforcement Learning for Multi-Modal\n Urban Traffic Control under Real-World limitations
• SPLADE v2: Sparse Lexical and Expansion Model for Information Retrieval
• Controllable Single-shot Animation Blending with Temporal Conditioning
• VQualA 2025 Challenge on Face Image Quality Assessment: Methods and Results
• Integral Transformer: Denoising Attention, Not Too Much Not Too Little
• Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic\n Circuits
• Multimodal Conditionality for Natural Language Generation
• Behavior From the Void: Unsupervised Active Pre-Training
• Learning ECG Representations via Poly-Window Contrastive Learning
• SparK: Query-Aware Unstructured Sparsity with Recoverable KV Cache Channel Pruning
• Automatic Graph Partitioning for Very Large-scale Deep Learning
• AutoHOOT: Automatic High-Order Optimization for Tensors
• Interactive Medical Image Segmentation with Self-Adaptive Confidence Calibration
• Efficient Identification of Critical Transitions via Flow Matching: A Scalable Generative Approach for Many-Body Systems
• TOAST: Fast and scalable auto-partitioning based on principled static analysis
• NCVX: A User-Friendly and Scalable Package for Nonconvex Optimization in Machine Learning
• Alias-Free Generative Adversarial Networks
• Convolutional Neural Network Pruning with Structural Redundancy Reduction
• Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules
• Tuning Random Generators: Property-Based Testing as Probabilistic Programming
• AFABench: A Generic Framework for Benchmarking Active Feature Acquisition
• DeepEmoNet: Building Machine Learning Models for Automatic Emotion Recognition in Human Speeches
• Beyond ReLU: Chebyshev-DQN for Enhanced Deep Q-Networks
• Great GATsBi: Hybrid, Multimodal, Trajectory Forecasting for Bicycles using Anticipation Mechanism
• BERT might be Overkill: A Tiny but Effective Biomedical Entity Linker based on Residual Convolutional Neural Networks
• Improving Joint Learning of Chest X-Ray and Radiology Report by Word Region Alignment
• GeMS: Efficient Gaussian Splatting for Extreme Motion Blur
• GiT: Graph Interactive Transformer for Vehicle Re-identification
• Machine Intelligence for Outcome Predictions of Trauma Patients During Emergency Department Care
• Data Distillation for Text Classification
• Lottery Jackpots Exist in Pre-trained Models
• An Expectation-Maximization Perspective on Federated Learning
• Skip-Connected Self-Recurrent Spiking Neural Networks with Joint Intrinsic Parameter and Synaptic Weight Training
• Agent with Warm Start and Adaptive Dynamic Termination for Plane Localization in 3D Ultrasound
• Data-Uncertainty Guided Multi-Phase Learning for Semi-Supervised Object Detection
• OccluNet: Spatio-Temporal Deep Learning for Occlusion Detection on DSA
• Predicting brain tumour enhancement from non-contrast MR imaging with artificial intelligence
• New Security Challenges on Machine Learning Inference Engine: Chip Cloning and Model Reverse Engineering
• Categorical Policies: Multimodal Policy Learning and Exploration in Continuous Control
• DIME-Net: A Dual-Illumination Adaptive Enhancement Network Based on Retinex and Mixture-of-Experts
• Explaining Data Anomalies over the NMSSM Parameter Space with Deep Learning Techniques
• Towards Fewer Annotations: Active Learning via Region Impurity and Prediction Uncertainty for Domain Adaptive Semantic Segmentation
• Learning to Learn the Macroscopic Fundamental Diagram using Physics-Informed and meta Machine Learning techniques
• Interactive Query Answering on Knowledge Graphs with Soft Entity Constraints
• MACTAS: Self-Attention-Based Module for Inter-Agent Communication in Multi-Agent Reinforcement Learning
• CRISP: Persistent Concept Unlearning via Sparse Autoencoders
• Personalized Subgraph Federated Learning with Sheaf Collaboration
• Weakly Supervised Anomaly Detection in Events with a Higgs Boson and Exotic Physics
• Prediction of Hospital Associated Infections During Continuous Hospital Stays
• Airborne acoustic emission enables sub-scanline keyhole porosity quantification and effective process characterization for metallic laser powder bed fusion
• Ranking over Regression for Bayesian Optimization and Molecule Selection
• SUB-Depth: Self-distillation and Uncertainty Boosting Self-supervised Monocular Depth Estimation
• Real-Time Reconstruction of 3D Bone Models via Very-Low-Dose Protocols
• Efficient neural encoding as revealed by bilingualism
• From ANN to BNN: Inferring Reionization Parameters using Uncertainty-aware Emulators of 21-cm Summaries
• GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks
• Fully Automated Segmentation of Fiber Bundles in Anatomic Tracing Data
• SparseMap: A Sparse Tensor Accelerator Framework Based on Evolution Strategy
• One-Class Intrusion Detection with Dynamic Graphs
• HRS: Hybrid Representation Framework with Scheduling Awareness for Time Series Forecasting in Crowdsourced Cloud-Edge Platforms
• LyricJam: A system for generating lyrics for live instrumental music
• DyCrowd: Towards Dynamic Crowd Reconstruction from a Large-scene Video
• Avaliação de eficiência na leitura: uma abordagem baseada em PLN
• A Self-Ensemble Inspired Approach for Effective Training of Binary-Weight Spiking Neural Networks
• aims-PAX: Parallel Active eXploration for the automated construction of Machine Learning Force Fields
• Lessons Learned Developing an Assembly System for WRS 2020 Assembly Challenge
• Improved Few-shot Segmentation by Redefinition of the Roles of Multi-level CNN Features
• Span Pointer Networks for Non-Autoregressive Task-Oriented Semantic Parsing
• PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds
• Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
• XAMT: Cross-Framework API Matching for Testing Deep Learning Libraries
• 3D Cardiac Anatomy Generation Using Mesh Latent Diffusion Models
• FLARE: Fast Low-rank Attention Routing Engine
• Causally-Guided Pairwise Transformer -- Towards Foundational Digital Twins in Process Industry
• OPTIC-ER: A Reinforcement Learning Framework for Real-Time Emergency Response and Equitable Resource Allocation in Underserved African Communities
• Polarization Reconfigurable Transmit-Receive Beam Alignment with Interpretable Transformer
• L-SR1: Learned Symmetric-Rank-One Preconditioning
• Dissecting Hessian: Understanding Common Structure of Hessian in Neural Networks
• DHG-Bench: A Comprehensive Benchmark for Deep Hypergraph Learning
• Learning Transferable 3D Adversarial Cloaks for Deep Trained Detectors
• Dense Regression Activation Maps For Lesion Segmentation in CT scans of COVID-19 patients
• Efficient Regional Memory Network for Video Object Segmentation
• Choice Set Optimization Under Discrete Choice Models of Group Decisions
• HFBTHO-AD: Differentiation of a nuclear energy density functional code
• Recent advances and applications of deep learning methods in materials science
• In Search of Lost Domain Generalization
• IC-Network: Efficient Structure for Convolutional Neural Networks
• Impact of Clinical Image Quality on Efficient Foundation Model Finetuning
• Persistence is All You Need -- A Topological Lens on Microstructural Characterization
• Limitation Learning: Catching Adverse Dialog with GAIL
• It's not a FAD: first results in using Flows for unsupervised Anomaly Detection at 40 MHz at the Large Hadron Collider
• Odd-One-Out Representation Learning
• Knowledge Base Completion for Constructing Problem-Oriented Medical\n Records
• Language models align with brain regions that represent concepts across modalities
• Self-Denoising Neural Networks for Few Shot Learning
• Why Settle for Just One? Extending EL++ Ontology Embeddings with Many-to-Many Relationships
• Learning to Predict Trustworthiness with Steep Slope Loss
• Learning with Algorithmic Supervision via Continuous Relaxations
• Unified Knowledge Distillation Framework: Fine-Grained Alignment and Geometric Relationship Preservation for Deep Face Recognition
• Minimizing Surrogate Losses for Decision-Focused Learning using Differentiable Optimization
• NeMo: A Neuron-Level Modularizing-While-Training Approach for Decomposing DNN Models
• Conformal Prediction Meets Long-tail Classification
• Master Thesis: Neural Sign Language Translation by Learning Tokenization
• Neural View Synthesis and Matching for Semi-Supervised Few-Shot Learning of 3D Pose
• Vax-a-Net: Training-time Defence Against Adversarial Patch Attacks
• Meta-learning Structure-Preserving Dynamics
• UAV-VL-R1: Generalizing Vision-Language Models via Supervised Fine-Tuning and Multi-Stage GRPO for UAV Visual Reasoning
• EMLIO: Minimizing I/O Latency and Energy Consumption for Large-Scale AI Training
• Learning Contextualised Cross-lingual Word Embeddings and Alignments for Extremely Low-Resource Languages Using Parallel Corpora
• Advances in Speech Separation: Techniques, Challenges, and Future Trends
• Interactive Recommender System via Knowledge Graph-enhanced Reinforcement Learning
• SemPT: Semantic Prompt Tuning for Vision-Language Models
• MirGuard: Towards a Robust Provenance-based Intrusion Detection System Against Graph Manipulation Attacks
• Physics-Informed Deep Contrast Source Inversion: A Unified Framework for Inverse Scattering Problems
• Projected Coupled Diffusion for Test-Time Constrained Joint Generation
• Virtual Sensing for Solder Layer Degradation and Temperature Monitoring in IGBT Modules
• Multi-Sample Anti-Aliasing and Constrained Optimization for 3D Gaussian Splatting
• Unpacking the Implicit Norm Dynamics of Sharpness-Aware Minimization in Tensorized Models
• Convolutional Hough Matching Networks for Robust and Efficient Visual Correspondence
• Meta-Metrics and Best Practices for System-Level Inference Performance Benchmarking
• ScaleCom: Scalable Sparsified Gradient Compression for Communication-Efficient Distributed Training
• A Comprehensive Evaluation framework of Alignment Techniques for LLMs
• Modern Neural Networks for Small Tabular Datasets: The New Default for Field-Scale Digital Soil Mapping?
• Embodied Tactile Perception of Soft Objects Properties
• Reconstruct high-resolution multi-focal plane images from a single 2D wide field image
• NEUBORN: The Neurodevelopmental Evolution framework Using BiOmechanical RemodelliNg
• Quest for a clinically relevant medical image segmentation metric: the definition and implementation of Medical Similarity Index
• A Worrying Analysis of Probabilistic Time-series Models for Sales Forecasting
• MoNet: Motion-based Point Cloud Prediction Network
• Swift for TensorFlow: A portable, flexible platform for deep learning
• MergeComp: A Compression Scheduler for Scalable Communication-Efficient Distributed Training
• Go Small and Similar: A Simple Output Decay Brings Better Performance
• WeatherPrompt: Multi-modality Representation Learning for All-Weather Drone Visual Geo-Localization
• Exploring the Equivalence of Closed-Set Generative and Real Data Augmentation in Image Classification
• Dynamical Alignment: A Principle for Adaptive Neural Computation
• HumanGenesis: Agent-Based Geometric and Generative Modeling for Synthetic Human Dynamics
• KonfAI: A Modular and Fully Configurable Framework for Deep Learning in Medical Imaging
• Inter-expert reliability in multi-field-of-view automatic drusen segmentation analysis using optical coherence tomography
• Relative Pose Regression with Pose Auto-Encoders: Enhancing Accuracy and Data Efficiency for Retail Applications
• Use of 1D-CNN for input data size reduction of LSTM in Hourly Rainfall-Runoff modeling
• Efficient estimates of optimal transport via low-dimensional embeddings
• Structured Reordering for Modeling Latent Alignments in Sequence Transduction
• Automated Charge Transition Detection in Quantum Dot Charge Stability Diagrams
• Neural quantum states for emitter dynamics in waveguide QED
• DiffPhysCam: Differentiable Physics-Based Camera Simulation for Inverse Rendering and Embodied AI
• Wavelet Mixture of Experts for Time Series Forecasting
• Visual Prompting for Robotic Manipulation with Annotation-Guided Pick-and-Place Using ACT
• WHAR Datasets: An Open Source Library for Wearable Human Activity Recognition
• Multi-level Collaborative Distillation Meets Global Workspace Model: A Unified Framework for OCIL
• Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation
• SignBERT: Pre-Training of Hand-Model-Aware Representation for Sign Language Recognition
• Unified and Semantically Grounded Domain Adaptation for Medical Image Segmentation
• Traffic Load-Aware Resource Management Strategy for Underwater Wireless Sensor Networks
• SegDAC: Visual Generalization in Reinforcement Learning via Dynamic Object Tokens
• Voice as a biomarker: exploratory analysis for benign and malignant vocal fold lesions
• Data Driven VRP: A Neural Network Model to Learn Hidden Preferences for\n VRP
• Generalized Operating Procedure for Deep Learning: an Unconstrained Optimal Design Perspective
• MonoPartNeRF:Human Reconstruction from Monocular Video via Part-Based Neural Radiance Fields
• Learning Generalizable and Efficient Image Watermarking via Hierarchical Two-Stage Optimization
• SelfHVD: Self-Supervised Handheld Video Deblurring
• Knowledge-guided machine learning for disentangling Pacific sea surface temperature variability across timescales
• Towards Efficient and Practical GPU Multitasking in the Era of LLM
• Segmentation of Satellite Imagery using U-Net Models for Land Cover Classification
• Point Cloud Registration using Representative Overlapping Points
• Dataset of Propaganda Techniques of the State-Sponsored Information Operation of the People's Republic of China
• Automated Decision-based Adversarial Attacks
• Muddling Labels for Regularization, a novel approach to generalization
• 3D Human Mesh Estimation from Single View RGBD
• Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning
• TRIDE: A Text-assisted Radar-Image weather-aware fusion network for Depth Estimation
• Meta Transition Adaptation for Robust Deep Learning with Noisy Labels
• Calculating the Projective Norm of higher-order tensors using a gradient descent algorithm
• Exploring Content and Social Connections of Fake News with Explainable Text and Graph Learning
• Vertex Features for Neural Global Illumination
• Deep Learning-Based Desikan-Killiany Parcellation of the Brain Using Diffusion MRI
• Forecasting Continuous Non-Conservative Dynamical Systems in SO(3)
• Grouped Speculative Decoding for Autoregressive Image Generation
• GAPNet: A Lightweight Framework for Image and Video Salient Object Detection via Granularity-Aware Paradigm
• ImageDDI: Image-enhanced Molecular Motif Sequence Representation for Drug-Drug Interaction Prediction
• Exploring Multimodal Diffusion Transformers for Enhanced Prompt-based Image Editing
• CNN Acceleration by Low-rank Approximation with Quantized Factors
• Designing with Deception: ML- and Covert Gate-Enhanced Camouflaging to Thwart IC Reverse Engineering
• Neural Logic Networks for Interpretable Classification
• Temporal User Profiling with LLMs: Balancing Short-Term and Long-Term Preferences for Recommendations
• Profiling Concurrent Vision Inference Workloads on NVIDIA Jetson -- Extended
• Intermediate-Task Transfer Learning with Pretrained Models for Natural Language Understanding: When and Why Does It Work?
• Cooperative Bi-path Metric for Few-shot Learning
• A Spin Glass Characterization of Neural Networks
• AutoAssert 1: A LoRA Fine-Tuned LLM Model for Efficient Automated Assertion Generation
• Enhancing Corrosion Resistance of Aluminum Alloys Through AI and ML Modeling
• SynMatch: Rethinking Consistency in Medical Image Segmentation with Sparse Annotations
• Progressive Point Cloud Deconvolution Generation Network
• Document Layout Analysis with Aesthetic-Guided Image Augmentation
• MCITlib: Multimodal Continual Instruction Tuning Library and Benchmark
• 3D Gaussian Representations with Motion Trajectory Field for Dynamic Scene Reconstruction
• Don't Wait, Just Weight: Improving Unsupervised Representations by Learning Goal-Driven Instance Weights
• Mode-Aware Non-Linear Tucker Autoencoder for Tensor-based Unsupervised Learning
• gpt-oss-120b &amp; gpt-oss-20b Model Card
• Towards Useful and Private Synthetic Omics: Community Benchmarking of Generative Models for Transcriptomics Data
• FrustraMPNN: An ultra-fast deep learning tool for proteome-scale analysis of deep mutational single-residue local energetic frustration in proteins
• Meta Auxiliary Learning for Facial Action Unit Detection
• IWA: Integrated Gradient based White-box Attacks for Fooling Deep Neural Networks
• Learning Models of Model Predictive Controllers using Gradient Data
• DISC: A Dynamic Shape Compiler for Machine Learning Workloads
• Taking A Closer Look at Synthesis: Fine-grained Attribute Analysis for Person Re-Identification
• An updated hybrid deep learning algorithm for identifying and locating\n primary vertices
• Being-ahead: Benchmarking and Exploring Accelerators for Hardware-Efficient AI Deployment
• MuSLCAT: Multi-Scale Multi-Level Convolutional Attention Transformer for Discriminative Music Modeling on Raw Waveforms
• DSConv: Dynamic Splitting Convolution for Pansharpening
• Federated Quantum Kernel-Based Long Short-term Memory for Human Activity Recognition
• MAHL: Multi-Agent LLM-Guided Hierarchical Chiplet Design with Adaptive Debugging
• Reverse Diffusion Sequential Monte Carlo Samplers
• Hybrid Physics-Machine Learning Models for Quantitative Electron Diffraction Refinements
• Recurrent Deep Differentiable Logic Gate Networks
• TorchSim: An efficient atomistic simulation engine in PyTorch
• What Voting Rules Actually Do: A Data-Driven Analysis of Multi-Winner Voting
• Comparative study of machine learning and statistical methods for automatic identification and quantification in γ-ray spectrometry
• Snowpark: Performant, Secure, User-Friendly Data Engineering and AI/ML Next To Your Data
• Integrating Vision Foundation Models with Reinforcement Learning for Enhanced Object Interaction
• Head Anchor Enhanced Detection and Association for Crowded Pedestrian Tracking
• RankArena: A Unified Platform for Evaluating Retrieval, Reranking and RAG with Human and LLM Feedback
• Input Invex Neural Network
• Learning Geometric-Aware Quadrature Rules for Functional Minimization
• MolSnap: Snap-Fast Molecular Generation with Latent Variational Mean Flow
• Artificial Intelligence-Based Classification of Spitz Tumors
• Multi-tracklet Tracking for Generic Targets with Adaptive Detection Clustering
• Learning from Similarity-Confidence and Confidence-Difference
• DQT: Dynamic Quantization Training via Dequantization-Free Nested Integer Arithmetic
• MetaDiT: Enabling Fine-grained Constraints in High-degree-of Freedom Metasurface Design
• TANGO: Graph Neural Dynamics via Learned Energy and Tangential Flows
• Aggregate Learning for Mixed Frequency Data
• Assessing Universal Relations for Rapidly Rotating Neutron Stars: Insights from an Interpretable Deep Learning Perspective
• I Think, Therefore I Am Under-Qualified? A Benchmark for Evaluating Linguistic Shibboleth Detection in LLM Hiring Evaluations
• ConfAgents: A Conformal-Guided Multi-Agent Framework for Cost-Efficient Medical Diagnosis
• Super Resolved Imaging with Adaptive Optics
• Case Studies of Generative Machine Learning Models for Dynamical Systems
• Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation
• GradSTL: Comprehensive Signal Temporal Logic for Neurosymbolic Reasoning and Learning
• T3Time: Tri-Modal Time Series Forecasting via Adaptive Multi-Head Alignment and Residual Fusion
• SplitGaussian: Reconstructing Dynamic Scenes via Visual Geometry Decomposition
• Bootstrap Deep Spectral Clustering with Optimal Transport
• Quantum Temporal Fusion Transformer
• A Comparative Survey of PyTorch vs TensorFlow for Deep Learning: Usability, Performance, and Deployment Trade-offs
• Deep learning framework for crater detection and identification on the Moon and Mars
• CoEmoGen: Towards Semantically-Coherent and Scalable Emotional Image Content Generation
• R2GenKG: Hierarchical Multi-modal Knowledge Graph for LLM-based Radiology Report Generation
• Exploring Layer-wise Information Effectiveness for Post-Training Quantization in Small Language Models
• Exploring Stability-Plasticity Trade-offs for Continual Named Entity Recognition
• Physics-Constrained Fine-Tuning of Flow-Matching Models for Generation and Inverse Problems
• Monocular Depth Estimation with Global-Aware Discretization and Local Context Modeling
• Description of CRESST-II and CRESST-III pulse shape data
• Urban In-Context Learning: Bridging Pretraining and Inference through Masked Diffusion for Urban Profiling
• Collab-Solver: Collaborative Solving Policy Learning for Mixed-Integer Linear Programming
• Adversarial Attention Perturbations for Large Object Detection Transformers
• Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images
• JustDense: Just using Dense instead of Sequence Mixer for Time Series analysis
• ASMR: Angular Support for Malfunctioning Client Resilience in Federated Learning
• GR-Gaussian: Graph-Based Radiative Gaussian Splatting for Sparse-View CT Reconstruction
• Forecasting When to Forecast: Accelerating Diffusion Models with Confidence-Gated Taylor
• Learning to Correspond Dynamical Systems
• Spatio-Temporal Sparsification for General Robust Graph Convolution Networks
• Zero-shot Compositional Action Recognition with Neural Logic Constraints
• Learning to Evolve: Bayesian-Guided Continual Knowledge Graph Embedding
• Proactive Disentangled Modeling of Trigger-Object Pairings for Backdoor Defense
• Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning
• Accumulative Poisoning Attacks on Real-time Data
• Single Point, Full Mask: Velocity-Guided Level Set Evolution for End-to-End Amodal Segmentation
• OpenMed NER: Open-Source, Domain-Adapted State-of-the-Art Transformers for Biomedical NER Across 12 Public Datasets
• Shape Distribution Matters: Shape-specific Mixture-of-Experts for Amodal Segmentation under Diverse Occlusions
• NS-Net: Decoupling CLIP Semantic Information through NULL-Space for Generalizable AI-Generated Image Detection
• OCSplats: Observation Completeness Quantification and Label Noise Separation in 3DGS
• ForenX: Towards Explainable AI-Generated Image Detection with Multimodal Large Language Models
• Multimodal Attention-Aware Fusion for Diagnosing Distal Myopathy: Evaluating Model Interpretability and Clinician Trust
• Explaining GNN Explanations with Edge Gradients
• HiPrune: Training-Free Visual Token Pruning via Hierarchical Attention in Vision-Language Models
• Video Color Grading via Look-Up Table Generation
• Foundations of Interpretable Models
• AniMer+: Unified Pose and Shape Estimation Across Mammalia and Aves via Family-Aware Transformer
• SDSNN: A Single-Timestep Spiking Neural Network with Self-Dropping Neuron and Bayesian Optimization
• Neighbor-Sampling Based Momentum Stochastic Methods for Training Graph Neural Networks
• Guided Depth Map Super-Resolution via Multi-Scale Fusion U-shaped Mamba Network
• Light-Weight Diffusion Multiplier and Uncertainty Quantification for Fourier Neural Operators
• Cross-Dataset Semantic Segmentation Performance Analysis: Unifying NIST Point Cloud City Datasets for 3D Deep Learning
• DACTYL: Diverse Adversarial Corpus of Texts Yielded from Large Language Models
• Quality-of-Service Aware LLM Routing for Edge Computing with Multiple Experts
• Tailoring: encoding inductive biases by optimizing unsupervised\n objectives at prediction time
• Deep learning enables city-wide climate projections of street-level heat stress
• Leveraging Operator Learning to Accelerate Convergence of the Preconditioned Conjugate Gradient Method
• Explainable Image Classification with Reduced Overconfidence for Tissue Characterisation
• DepMicroDiff: Diffusion-Based Dependency-Aware Multimodal Imputation for Microbiome Data
• Hierarchical Message-Passing Policies for Multi-Agent Reinforcement Learning
• 3D-MOOD: Lifting 2D to 3D for Monocular Open-Set Object Detection
• Stable-Sim2Real: Exploring Simulation of Real-Captured 3D Data with Two-Stage Depth Diffusion
• Machine learning and machine learned prediction in chest X-ray images
• Long-Range Transformers for Dynamic Spatiotemporal Forecasting
• Classification with Rejection Based on Cost-sensitive Classification
• FMIP: Joint Continuous-Integer Flow For Mixed-Integer Linear Programming
• Designing Dynamic Pricing for Bike-sharing Systems via Differentiable Agent-based Simulation
• Simulation-based inference for Precision Neutrino Physics through Neural Monte Carlo tuning
• SequenceLayers: Sequence Processing and Streaming Neural Networks Made Easy
• From LLMs to Edge: Parameter-Efficient Fine-Tuning on Edge Devices
• VMatcher: State-Space Semi-Dense Local Feature Matching
• Graph Lineages and Skeletal Graph Products
• MRpro - open PyTorch-based MR reconstruction and processing package
• Predicting Terrain Mechanical Properties in Sight for Planetary Rovers with Semantic Clues
• DSXplore: Optimizing Convolutional Neural Networks via Sliding-Channel Convolutions
• KLLM: Fast LLM Inference with K-Means Quantization
• TR-PTS: Task-Relevant Parameter and Token Selection for Efficient Tuning
• The Devil is in the Detail: Simple Tricks Improve Systematic\n Generalization of Transformers
• Minimizing Oracle-Structured Composite Functions
• SPARE3D: A Dataset for SPAtial REasoning on Three-View Line Drawings
• Deep Implicit Coordination Graphs for Multi-agent Reinforcement Learning
• Secure Domain Adaptation with Multiple Sources
• Orthonormal Product Quantization Network for Scalable Face Image Retrieval
• DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors
• Image-Guided Shape-from-Template Using Mesh Inextensibility Constraints
• A holomorphic Kolmogorov-Arnold network framework for solving elliptic problems on arbitrary 2D domains
• trAIce3D: A Prompt-Driven Transformer Based U-Net for Semantic Segmentation of Microglial Cells from Large-Scale 3D Microscopy Images
• Type-driven Neural Programming by Example
• End-to-End Learning of Flowchart Grounded Task-Oriented Dialogs
• Randomized Overdrive Neural Networks
• Robust Deepfake Detection for Electronic Know Your Customer Systems Using Registered Images
• HGCN(O): A Self-Tuning GCN HyperModel Toolkit for Outcome Prediction in Event-Sequence Data
• FDNAS: Improving Data Privacy and Model Diversity in AutoML
• Tensor (machine learning) [wikipedia]

Discussions

• PyTorch: An Imperative Style, High-Performance Deep Learning Library [pdf] [hn, 113 points, 16 comments]

Related

• PyTorchPipe: a framework for rapid prototyping of pipelines combining language and vision
• Torch.fx: Practical Program Capture and Transformation for Deep Learning in Python
• TorchBench: Benchmarking PyTorch with High API Surface Coverage
• OpTorch: Optimized deep learning architectures for resource limited environments
• PyTorch Tabular: A Framework for Deep Learning with Tabular Data