GNN Explainability

GNN: Graph Neural Network

Ask about this topic

Overview

Graph neural network explainability seeks to show which node attributes, connections, subgraphs, or concepts drive predictions, including in temporal and heterogeneous graphs. Post-hoc approaches use masks, gradients, Shapley values, and counterfactual edits, while intrinsic methods build explanations into models through concept bottlenecks, sparse additive scores, or symbolic rules. Researchers evaluate explanations with fidelity, stability, sparsity, and ground-truth motifs on synthetic benchmarks as well as molecular, biological, and social-network tasks. A central challenge is that explanations can change with the explainer, graph structure, or evaluation metric, and a faithful account of model behavior does not by itself establish a causal or scientifically valid mechanism.

Latest papers

CardsList

82 papers10 papers in the last four weeks, up 67% on the four weeks before. 0.1% of all new papers.

  1. GNN-VIEW: Visual Inspection and Explanation of Graph-based Earth-System Models

    Oct 26, 2026Alessandro Lucatelli, Ferdinand Unterhuber, Katharina Krenzer +4GNN ExplainabilityGraph Neural Networks

    ETH Zürich, Switzerland

  2. Fragment-Level Graph Encoding for Interpretable Molecular Property Prediction

    Oct 21, 2026Rafael Lopes Almeida, Ina PöhnerMolecular Property PredictionGNN Explainability

    School of Pharmacy, University of Eastern Finland, Kuopio, Finland

  3. Declarative Explanations for Graph Neural Networks: A Demonstration

    Oct 19, 2026Cibo Yu, Haitong Tang, Yinghui Wu +4GNN ExplainabilityGraph Neural Networks

    Zhejiang University, Hangzhou, China · Case Western Reserve University, Cleveland, USA · Bowling Green State University, Bowling Green, USA

  4. Graph Additive Models disentangling feature, topology, and interactions

    Oct 14, 2026Clara Alvarez-Rodriguez, Steve Azzolin, Vincenzo Marco De Luca +3Graph Neural NetworksGNN Explainability

    The Arctic University of Norway

  5. xLightGCN: Explainable Graph Neural Network Based Recommendation Systems with Most Influential Peers

    Oct 14, 2026Calder Johnson, Robin CohenExplainable RecommendationRecommender Systems

    Computer Science, University of Waterloo, Canada

  6. Does the Explanation Drift Before the Guarantee? Label-Free Early Warning on Evolving Graphs

    Oct 12, 2026Tomasz Szandała, Damian RaczkowskiDistribution Shift DetectionGNN Explainability

    Faculty of Information and Communication Technology, Wroclaw University of Science and Technology · independent researcher

  7. MPGE: A Multi-Perspective Graph Explainer for Molecular Classification Explanation

    Oct 8, 2026Mahtab SarvmailiGraph Counterfactual ExplanationsGNN Explainability

    Dalhousie University Halifax, Nova Scotia, Canada

  8. Beyond Trained Models: Compiling GNNs for a Sound Explainer Benchmark

    Oct 2, 2026Steve Azzolin, Francesco Paolo Nerini, Stefano Teso +4GNN ExplainabilityExplainability Evaluation

    University of Trento Trento, Italy

  9. WOMBAT: Whitebox Oracle for Molecular Benchmarking and Attribution Testing

    Sep 30, 2026Dominik Matuszek, Bartosz Zieliński, Tomasz Danel +1Gradient-Based AttributionExplainability Evaluation

    Jagiellonian University, Faculty of Mathematics and Computer Science · Jagiellonian University, Doctoral School of Exact and Natural Sciences · Jagiellonian University, Jagiellonian Center for Artificial Intelligence +2

  10. Where Privacy Belongs: Placement Diagnosis and Certified Selection for Private Counterfactual Explanations on Graphs

    Sep 29, 2026Yuxiang Yao, Zijun ZhaoCounterfactual ExplanationsGNN Explainability

    Project Management Department, Research Center, China Life Insurance Company Ltd., Beijing, China · School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China

  11. RACE: Relation-Level Counterfactual Explanations for Heterogeneous Graph Neural Networks

    Sep 29, 2026Yuxiang Yao, Zijun ZhaoGraph Counterfactual ExplanationsGNN Explainability

    Project Management Department, Research Center, China Life Insurance Company Ltd. · School of Computer Science and Technology, Beijing Institute of Technology

  12. HARMONIA: Interpretable Graph Learning through Mixtures of Neural Bases

    Sep 27, 2026Quan D. Bui, Nguyen Do, An Nguyen Dang +3Graph Representation LearningGNN Explainability

    Center for AI Research, VinUniversity, Hanoi, Vietnam · University of Florida, Gainesville, USA · Posts and Telecommunications Institute of Technology, Hanoi, Vietnam +1

  13. Controllable GNN Explanations via Multi-Metric Preference Selection

    Sep 26, 2026Rachit Verma, Yashraj J. Deshmukh, Anirban DasguptaGNN ExplainabilityExplanation Stability

    Indian Institute of Technology Gandhinagar

  14. Temporal Generalization and Explanation Stability of Control Flow Graph Neural Networks for Malware Detection

    Sep 21, 2026Md. Asif Sajeed, Md. Nazrul Islam Mondal, Md Ashraful Hossen AkashGraph Neural NetworksDistribution Shift

    Department of Computer Science and Engineering, Rajshahi University of Engineering & Technology, Rajshahi, 6204, Bangladesh

  15. An Accurate and Interpretable Hyper Graph Neural Network for GBM Survival Prediction

    Sep 19, 2026Mushahid IntesumCancer Survival PredictionMultimodal Survival Analysis

  16. A Comparative Study of Counterfactual Explainers for Graph Neural Networks Enabling Multiple Types of Graph Edit

    Sep 4, 2026Maria Myrto Villia, Filippos Gouidis, Theodore Patkos +1Counterfactual ExplanationsGraph Counterfactual Explanations

    Institute of Computer Science, Foundation for Research and Technology - Hellas (FORTH) · Computer Science Department, University of Crete Greece

  17. MolLedger: An Additive Graph Neural Network with Chemically Grounded ADME Attributions

    Aug 31, 2026Christina X. JiFeature AttributionGraph Neural Networks

    Hamilton College

  18. Self-Explainable Multi-Label Graph Neural Network for Correlated Evidence Attribution

    Aug 27, 2026Yingqi Feng, Yufei Tang, Min Shi +1GNN ExplainabilityMulti-Label Classification

    Department of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL, USA · School of Computing and Informatics, University of Louisiana at Lafayette, Lafayette, LA, USA

  19. Generating Intervention Hypotheses using Explainable Explanations on Graphs: G2I, a Two-Stage Greedy Framework

    Aug 24, 2026Mulin Tian, Ajitesh SrivastavaGNN ExplainabilityGraph Counterfactual Explanations

    University of Southern California Los Angeles, California, USA · Northeastern University Charlotte, North Carolina, USA

  20. From Abductive Explanations to Global Logical Rules for Node Classification in SGCs

    Aug 17, 2026Bryan Lima Cavalcante, Thiago Alves RochaGNN ExplainabilityGraph Neural Networks

    Instituto Federal do Cear´a (IFCE) Brazil

  21. NICE: Scale-Stable Perturbations for Graph Neural Network Explanations via Noise Corruption

    Aug 17, 2026Ziluowen Luo, Jun Yin, Ruochen Liu +4GNN ExplainabilityPerturbation-Based Feature Attribution

    Central South University · Hong Kong Polytechnic University · Griffith University

  22. Amortised Post-Hoc Explanation with Exact Preservation for Dynamic Graph Anomaly Detectors

    Aug 16, 2026Iyad Assaad Nekka, Hamida Seba, Walid Khaled Hidouci +1GNN ExplainabilityInterpretable Anomaly Detection

    LCSI Laboratory, National higher School of Computer Science (ESI), Algiers, Algeria · Université Claude Bernard Lyon, Lyon, France

  23. Structure-Guided Spatiotemporal Attention Graph Neural Network for Traffic Flow Prediction

    Aug 14, 2026Xuanmian He, Can Li, Wanjing MaTraffic Flow EstimationSpatiotemporal Forecasting

    Department of Civil and Environmental Engineering, University of California, Berkeley, United States · Key Laboratory of Road and Traffic Engineering of the Ministry of Education, Tongji University, Shanghai 201804, China

  24. Overcoming Shortcut Learning in Graph Neural Networks through Active Explanation Guidance

    Aug 14, 2026Taraneh Younesian, Steve Azzolin, Antonio Longa +3Graph Neural NetworksShortcut Learning

    VU Amsterdam, Netherlands · University of Trento, Italy · UiT The Arctic University of Norway, Norway

  25. Dual Spatial-Temporal Attribution: Architecture-Aligned Post-Hoc Explainability for Recurrent Graph Anomaly Detection

    Aug 12, 2026Iyad Assaad Nekka, Hamida Seba, Khaled Walid Hidouci +1Gradient-Based AttributionInterpretable Anomaly Detection

    LCSI Laboratory, National Higher School of Computer Science (ESI), Algiers, Algeria · Université Claude Bernard Lyon 1, Lyon, France

  26. Faithful, Sufficient and Understandable: Rethinking Graph Counterfactual Explanations via Discrete Diffusion Inversion

    Aug 12, 2026David Bechtoldt, Sidney BenderCounterfactual EvaluationCounterfactual Explanations

    Machine Learning Group, TU Berlin, 10587 Berlin, Germany · BIFOLD–Berlin Institute for the Foundations of Learning and Data, Berlin, Germany

  27. An Explainable GNN Framework for Component-Level Anomaly Diagnosis

    Aug 10, 2026Sena Ozgunay, Louise Trav{é}-Massuy{è}s, Jean-Michel Loubes +1Industrial Anomaly DetectionRoot Cause Analysis

    Comue de Toulouse, LAAS-DISCO, IMT · Comue de Toulouse, LAAS-DISCO · REGALIA, IMT +1

  28. Patients-like-me: A Variational LM--GNN Framework for Explainable Clinical Prediction

    Aug 4, 2026Xinyu Wang, Yixuan Li, Hanwei Wu +4Graph Neural NetworksClinical Outcome Prediction

    McGill University · Université de Montréal · Georgia State University +1

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

    Jul 28, 2026Xin Xu, Siru Tao, Kaizhen TanGradient-Based AttributionSymmetry Breaking

    Carnegie Mellon University

  30. Learned, Relied Upon, or Necessary? Separating Checkpoint Dependence from Task-Level Value in Sheaf GNNs

    Jul 28, 2026Yi LiuSheaf Neural NetworksSheaf Theory

    School of Astronomy and Space Science, University of Science and Technology of China