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

#graph neural networks

82 results
q-fin.RM2026

No Data Is Not No Risk: Visibility Aware Graph-Based Inference of Business Conduct Risk

Tsuyoshi Iwata, Johannes Laurmaa, Ryohei Hisano

The paper proposes a visibility‑aware graph neural network that leverages corporate ownership relationships to predict business conduct risk for firms lacking recorded incidents, t…

#graph neural networks#positive-unlabeled learning#business conduct risk#corporate ownership networks
cs.CL2026

AtmosERC: Modeling Dialogue-Level Affective Atmosphere for Emotion Recognition in Conversation

Weijie Feng, Tongwei Zhang, Binbin Liu +1

The paper introduces AtmosERC, a graph-based framework that captures a dialogue-level affective atmosphere to improve emotion recognition in conversations, providing both lightweig…

#emotion recognition#conversation analysis#graph neural networks#affective computing
cs.LG2026

Neural Architecture Search for Traffic Prediction: A Survey of Methods, Challenges, and Future Directions

Truong Giang Vu, Li Yang, Richard W. Pazzi

The paper surveys how neural architecture search techniques are used to automatically design deep learning models for traffic prediction, reviewing gradient‑based, evolutionary, an…

#traffic prediction#neural architecture search#graph neural networks#spatiotemporal modeling
eess.SP2026

Multi-User Localization via Active Sensing with Electromagnetically Reconfigurable Antennas

Ruizhi Zhang, Yuchen Zhang, Ying Zhang +1

The paper proposes an active‑sensing framework for multi‑user uplink localization that adaptively configures shared electromagnetically reconfigurable antennas using past pilot obs…

#multi-user localization#active sensing#reconfigurable antennas#graph neural networks
cs.LG2026

Schreier-Coset Graph Rewiring

Aryan Mishra, Randy Martinez, Lizhen Lin

The paper proposes Schreier-Coset Graph Rewiring, a group‑theoretic method that augments a graph with a Schreier‑Coset graph to reduce over‑squashing in graph neural networks while…

#graph neural networks#graph rewiring#over-squashing#spectral graph theory
cs.LG2026

THGFM: Dual-Branch Temporal Heterogeneous Graph Fusion Model

Yixin Peng, Diego Collarana, Er Jin +1

The paper introduces THGFM, a dual-branch graph transformer model that jointly handles structural heterogeneity and temporal dynamics in heterogeneous graphs using shared and speci…

#temporal heterogeneous graphs#graph neural networks#attention mechanisms#graph transformers
cs.AI2026

Leveraging Trajectory Graphs for Pre-Execution Error Diagnosis in Agentic LLM Systems

Xu Zheng, Zhuomin Chen, Chaohao Lin +4

The paper introduces Trajectory Graph Copilot, a framework that builds probabilistic graphs of past agent trajectories and uses a graph neural network to flag potentially erroneous…

#large language models#agentic systems#trajectory analysis#graph neural networks
cs.LG2026

Automorphism-Induced Non-Canonicity in Top-k Explanations of Graph Neural Networks

Xin Xu, Siru Tao, Kaizhen Tan

The paper shows that gradient‑based explainers for graph neural networks can produce arbitrary top‑k edge explanations when the input graph has nontrivial automorphisms, and provid…

#graph neural networks#explainability#graph automorphisms#top‑k explanations
cs.CR2026

Learning the Word Problem: Geodesic Lengths and Cryptographic Applications

Elisabeth Fink

The paper presents WPNet, a graph neural network that learns to solve the Word Problem and predict geodesic lengths in certain infinite groups, and shows how this can be used to at…

#word problem#group theory#graph neural networks#cryptographic attacks
cs.AI2026

TRWH: A Text-Driven Random Walk Heterogeneous GNN for Semantic-Aware Sparse Recommendation

He Ma, Chen Liu

The paper introduces TRWH, a framework that combines large language model‑generated text profiles with heterogeneous graph neural networks and random‑walk augmentation to improve r…

#graph neural networks#large language models#sparse recommendation#heterogeneous graphs
cs.IR2026

MARS: Multi-Agent Re-ranking for Repeat-Order Food Delivery Recommendation

Jiahao Tian, Zhenkai Wang

The paper presents MARS, a modular multi‑agent framework that combines large language model reasoning with lightweight collaborative retrieval and geospatial filtering to re‑rank r…

#food delivery recommendation#multi-agent re-ranking#large language models#collaborative filtering
physics.chem-ph2026

Accelerated descriptor-free path sampling for protein-ligand binding kinetics

Simon M. Lichtinger, Roberto Covino

The paper introduces a descriptor‑free path‑sampling method that uses an equivariant graph neural network to model the committor and a static bias potential to accelerate convergen…

#protein-ligand binding#kinetics#path sampling#graph neural networks
cs.SE2026

Towards Predicting Multi-Vulnerability Attack Chains in Software Supply Chains from Software Bill of Materials Graphs

Laura Baird, Armin Moin

The paper introduces a graph‑learning method that uses Software Bill of Materials (SBOM) graphs to predict cascades of multiple vulnerabilities in software supply chains, combining…

#software supply chain security#sbom analysis#vulnerability attack chains#graph neural networks
cs.CL2026

Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution

Zlata Kikteva, Artur Romazanov, Annette Hautli-Janisz +1

The paper introduces a method that extracts reasoning structures from LLM‑generated text and uses a graph neural network to attribute authorship, showing higher robustness to parap…

#authorship attribution#large language models#reasoning graphs#graph neural networks
cs.RO2026

Octopus-inspired Distributed Control for Soft Robotic Arms: A Graph Neural Network-Based Attention Policy with Environmental Interaction

Linxin Hou, Qirui Wu, Zhihang Qin +2

The paper introduces SoftGM, a graph‑neural‑network based distributed control system for segmented soft robotic arms that learns to reach targets while discovering obstacles online…

#soft robotics#distributed control#graph neural networks#multi-agent reinforcement learning
cs.CV2026

DAPGNet: Dynamic Adaptive Physics-Guided Graph Diffusion Network for Hyperspectral Image Classification

Pengkun Wang, Weijia Cao, Ning Wang +1

The paper proposes DAPGNet, a graph diffusion network that incorporates physical priors from contiguous spectral bands to improve hyperspectral image classification, using adaptive…

#hyperspectral image classification#graph neural networks#physics‑guided learning#spectral‑spatial modeling
cs.AI2026

A short review on the maximum clique problem algorithms with classical, AI, and quantum methods

Raffaele Marino, Lorenzo Buffoni, Bogdan Zavalnij

The paper surveys algorithms for solving the maximum clique problem, covering classical exact and heuristic methods as well as recent graph neural network and quantum computing app…

#maximum clique problem#graph algorithms#classical algorithms#graph neural networks
q-bio.BM2026

Training a force field for proteins and small molecules from scratch

Alexandre Blanco-González, Thea K Schulze, Evianne Rovers +1

The paper introduces Garnet, a graph neural network that learns force field parameters for proteins and small molecules directly from quantum mechanical and experimental data, achi…

#force fields#graph neural networks#molecular dynamics#protein modeling
cond-mat.mtrl-sci2026

SevenNet-Polar for MultiTask Prediction of Energy, Forces, Stress, and Born Effective Charges: Development and Application to ZrO$_2$, Li$_3$PO$_4$, and Perovskites

Anh Khoa Augustin Lu, Shungo Arai, Yutack Park +3

The paper introduces SevenNet-Polar, an equivariant graph neural network that simultaneously predicts energy, forces, stress, and Born effective charge tensors with high accuracy,…

#graph neural networks#born effective charge#multitask learning#molecular dynamics
cs.LG2026

Grad2Fair: A Gradient-driven Approach for Graph Fairness without Demographics

Yuchang Zhu, Zezhong Xie, Huizhe Zhang +4

The paper introduces Grad2Fair, a method that uses gradient information to detect and reduce group bias in graph neural networks without requiring demographic attributes, achieving…

#graph neural networks#fairness#group bias#gradient-based methods
cs.LG2026

MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model

Charilaos Papaioannou, Ioannis Tsantilas, Dimitris Giannakakos +9

The paper presents MxGPS, a multiplex graph transformer that jointly trains on static state estimation and AC power flow to avoid topology overfitting, achieving robust zero‑shot p…

#graph neural networks#power grid modeling#multi-task learning#graph transformers
cs.LG2026

NeuroGRIP: Retrieval-Augmented Graph Refinement for Knowledge-Grounded EEG Seizure Diagnosis

Lincan Li, Zheng Chen, Yushun Dong

NeuroGRIP is a framework that refines EEG-based graph neural network predictions for seizure diagnosis by retrieving and integrating clinical knowledge from a domain-specific knowl…

#eeg seizure detection#graph neural networks#knowledge graphs#retrieval-augmented reasoning
physics.ao-ph2026

OCELOT: Direct Atmospheric Forecasting from Heterogeneous Earth Observations Using a Graph-Transformer Hybrid Model

Azadeh Gholoubi, Ronald McLaren, Mu-Chieh Ko +7

The paper introduces OCELOT, a machine‑learning system that directly forecasts atmospheric observations up to 12 hours ahead by processing heterogeneous satellite and in‑situ data…

#weather forecasting#machine learning#graph neural networks#satellite observations
cs.LG2026

LATTICE: Graph Self-Supervised Learning for Multimodal Spatial Omics Integration

Jagan Mohan Reddy Dwarampudi, Veena Kochat, Suresh Satpati +3

LATTICE is a graph-based self‑supervised framework that learns spot‑level embeddings by integrating multimodal spatial omics data (RNA, ATAC, CUT&Tag) using a TransformerConv encod…

#spatial omics#multimodal integration#graph neural networks#self-supervised learning