NewEvery arXiv paper, its researchers & institutions — mapped.
the archive

#graph neural networks

82 results
cs.AI2026

CausalGraphX: A Counterfactual Graph Neural Network Framework for Explainable Systemic Risk Assessment

Rabimba Karanjai, Hemanth Madhavarao, Lei Xu +1

The paper presents CausalGraphX, a framework that combines graph neural networks with counterfactual reasoning to predict and explain systemic risk in financial networks, offering…

#systemic risk#graph neural networks#counterfactual reasoning#explainable ai
q-bio.QM2026

A Machine Learning Benchmarking Framework for Lipid Nanoparticle Transfection Efficiency Prediction

Asal Mehradfar, Mohammad Shahab Sepehri, Jose Miguel Hernandez-Lobato +4

The paper introduces a standardized machine‑learning benchmarking framework for predicting lipid‑nanoparticle transfection efficiency from ionizable lipid structures, evaluating ma…

#lipid nanoparticles#transfection efficiency#molecular representation#benchmarking
cs.LG2026

Low-Latency Relay Selection in NR-V2X Vehicular Communications via Graph Isomorphism Networks with Edge Features

Giambattista Amati, Federica Mangiatordi, Emiliano Pallotti +3

The paper proposes a graph‑neural‑network approach (GINE) to select relay nodes for low‑latency NR‑V2X communications, using edge features and an offline MILP oracle to train the m…

#vehicular communications#relay selection#graph neural networks#edge features
physics.flu-dyn2026

Evaluation of State-of-the-Art Deep Learning Architectures for Aerodynamical Predictions

Jan Scherz, Derrick Hines, Philipp Bekemeyer

The paper benchmarks four modern deep learning operator learning models as surrogate solvers for aerodynamic predictions, evaluating their ability to predict surface pressure on 2D…

#operator learning#surrogate modeling#aerodynamics#deep learning
cs.LG2026

NodeImport: Imbalanced Node Classification with Node Importance Assessment

Nan Chen, Zemin Liu, Bryan Hooi +3

The paper proposes NodeImport, a framework that assesses node importance using a balanced meta-set to dynamically select valuable labeled, unlabeled, and synthetic nodes, improving…

#node classification#class imbalance#graph neural networks#node importance assessment
cs.LG2026

Gauge-Invariant, Parameter-Insensitive Regularization for Potential Recovery from Flow on Directed Graphs

Mohammad Forouhesh

The paper proposes a gauge‑invariant regularization based on the graph Dirichlet energy to recover latent potentials from flow data on directed graphs, achieving stable estimates a…

#graph signal processing#inverse problems#regularization#directed graphs
cs.CE2026

Accelerating gas-network feasibility screening with a physics-informed graph neural network surrogate

Dongrui Jiang, Jochen Garcke, Okan Akca +4

The paper presents a physics‑informed graph neural network that quickly predicts steady‑state pressures and flows in large gas networks while enforcing mass‑balance, enabling rapid…

#gas networks#graph neural networks#physics-informed learning#feasibility screening
cs.LG2026

Implementations of Quantum and Classical Topology-Aligned Architectures for Molecular Property Prediction

James T. Pegg, Hubert Okadome Valencia, Ronin Wu

The paper introduces topology‑aligned architectures that map molecular bond graphs onto learnable units, implemented as a variational quantum circuit (Iso‑QGNN) and a matching clas…

#quantum machine learning#graph neural networks#molecular property prediction#parameter efficiency
cs.CV2026

DreamSat-Pose: Spacecraft Pose Estimation from Single-View 3D Reconstructions and Learned 2D-3D Feature Matching

Josiane Uwumukiza, Jocelyn Zhao, Giovanni Lavezzi +5

The paper introduces a single-shot framework that reconstructs a 3D shape of an unknown spacecraft from one image and then estimates its 6-DoF pose by learning dense 2D‑3D correspo…

#spacecraft pose estimation#single-view 3d reconstruction#2d-3d feature matching#transformer networks
cs.LG2026

CDS: Counterfactual Directionality Score for Structured Interventions in Spatial Graphs

Humaira Anzum, Md Ishtyaq Mahmud, Jagan Mohan Reddy Dwarampudi +1

The paper proposes a framework that uses structured counterfactual interventions on graph models to quantify directional influence between different node types, introducing the Cou…

#graph neural networks#counterfactual inference#spatial graphs#directional influence
cs.NI2026

GNN-based Online Beamforming Design for HAPS-Assisted NTN

Lavanya S S Anjapuli, Animesh Yadav, Halim Yanikomeroglu

The paper proposes using a high‑altitude platform station (HAPS) to relay data for cell‑edge users and designs beamforming vectors at both the terrestrial base station and HAPS via…

#high-altitude platform stations#beamforming#graph neural networks#energy efficiency
cs.LG2026

Variational Mixture of Graph Neural Experts for Alzheimer's Disease Recognition across Frequency Bands in EEG Brain Networks

Jun-En Ding, Anna Zilverstand, Shihao Yang +2

The paper introduces VMoGE, a variational mixture-of-experts model that uses graph neural networks to analyze EEG connectivity across multiple frequency bands for distinguishing Al…

#eeg analysis#graph neural networks#dementia diagnosis#variational inference
cs.RO2026

GNN-DIP: Neural Corridor Selection for Decomposition-Based Motion Planning

Peng Xie, Yanliang Huang, Wenyuan Wu +1

The paper introduces GNN-DIP, a method that uses a graph neural network to score portals in a cell adjacency graph, guiding corridor selection for decomposition-based motion planni…

#motion planning#narrow passages#graph neural networks#decomposition
cs.SI2026

Beyond Parents? Prediction Gaps in University Completion Using Population-Scale Networks and Flexible Machine Learning

Javier Garcia-Bernardo, Eva Jaspers, Weverthon Machado +2

The paper investigates how much children’s university completion can be predicted from broader social contexts beyond parental background, using large administrative datasets and c…

#educational attainment#social networks#machine learning#graph neural networks
cs.LG2026

Explaining Temporal Graph Neural Networks via Feature-induced Information Flow

Ping Xiong, Thomas Schnake, Klaus-Robert Müller +1

The paper introduces an attribution method that explains temporal graph neural networks by quantifying information flow through both event embeddings and event-induced variables, i…

#temporal graphs#explainability#graph neural networks#information flow
cs.CR2026

StableAML: Machine Learning for Behavioral Wallet Detection in Stablecoin Anti-Money Laundering on Ethereum

Luciano Juvinski, Haochen Li, Alessio Brini

The paper presents a machine‑learning system that uses tree‑ensemble models to identify suspicious stablecoin wallets on Ethereum, achieving higher detection performance than graph…

#stablecoins#anti-money laundering#wallet detection#ethereum
eess.SP2026

Physics-Informed Structure Anchoring With Capture-Aware Prototype Calibration for Cross-Environment RF Fingerprinting

Fengchong Yao, Jianbing Li, Qing Liu +4

The paper introduces PISA-CAPC, a method that uses physics‑informed graph representations of antenna topology and carrier‑frequency‑offset conditioning, together with unlabeled pro…

#radio frequency fingerprinting#cross-environment adaptation#graph neural networks#prototype calibration
eess.SP2026

Positional Attention-based Graph Neural Network for Learning Permutation Non-equivariant Wireless Policies

Baichuan Zhao, Chenyang Yang, Jianyu Zhao +1

The paper introduces a positional attention‑based graph neural network that can learn wireless policies which are not permutation equivariant, improving channel estimation and end‑…

#graph neural networks#permutation non-equivariance#positional attention#channel estimation
cs.LG2026

RF-Informed Graph Neural Networks for Accurate and Data-Efficient Circuit Performance Prediction

Anahita Asadi, Leonid Popryho, Inna Partin-Vaisband

The paper introduces a graph neural network framework that uses RF-specific feature encoding to predict performance metrics of various active RF circuits with high accuracy and low…

#graph neural networks#rf circuit modeling#performance prediction#data-efficient learning
cs.AI2026

Improving Molecular Property Prediction in Small Language Models Using Graph-based Tools

Konstantinos Bougiatiotis, Dimitrios Kelesis, Georgios Paliouras

The paper proposes a Context‑Augmented Prompting framework that lets small language models query a graph neural network expert for structural hints and explanatory subgraphs, impro…

#small language models#molecular property prediction#graph neural networks#prompt engineering
cs.LG2026

STKAN: Kolmogorov-Arnold Networks for Spatio-Temporal Forecasting

Sicong Lai, Yuehong Hu, Siru Zhong +3

The paper introduces STKAN, a spatio‑temporal forecasting model that uses Kolmogorov‑Arnold network modules with Taylor‑polynomial approximations for spatial and temporal token mix…

#spatio-temporal forecasting#traffic prediction#graph neural networks#kolmogorov-arnold networks
cs.LG2026

EMAGN: Efficient Multi-Attention Graph Network via Learned Clustering for Scalable Traffic Forecasting

Mingxing Xu, Rakesh Chowdary Machineni, Ke Liu +7

The paper introduces EMAGN, a graph neural network that uses learned clustering to linearize multi-head attention, enabling scalable traffic forecasting with reduced computation an…

#traffic forecasting#graph neural networks#attention mechanisms#efficient inference
cs.LG2026

A multi-scale feature enhanced graph neural network for fluid dynamics prediction in complex geometries

Li Xiao, Tianyu Li, Yiye Zou +2

The paper introduces ME-GNN, a multi‑scale feature enhanced graph neural network that combines two‑step message passing, an Attention U‑Net, and K‑hop sampling to predict fluid flo…

#graph neural networks#fluid dynamics prediction#multi‑scale features#complex geometries
eess.SP2026

Learning-based Multiuser Beamforming for Holographic MIMO~Systems

Shiyong Chen, Shengqian Han

The paper proposes a learning‑based approach using a gradient‑based graph neural network to design multi‑user beamformers for holographic MIMO systems, achieving higher spectral ef…

#holographic mimo#beamforming#graph neural networks#multiuser communications