Skip to content

gongchenghua/Papers-Graphs-with-Heterophily

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

78 Commits
 
 
 
 

Repository files navigation

A Survey on Learning from Graphs with Heterophily: Progress and Future

PRs Welcome Awesome Stars Forks

This repository contains a list of the relevant resources on learning from graphs with heterophily. We categorize the following papers based on their published years. We will try our best to continuously maintain this repository in real time. If you found any error or any missed paper, please don't hesitate to open issues.

Literature Overview

In this survey, we summarize over 500 high-quality papers published in top conferences or journals over the past 5 years including but not limited to ICML, NeurIPS, ICLR, KDD, WWW, AAAI, IJCAI, ICDE, TPAMI, TKDE etc. To catch up with the frontiers, some latest works on OpenReview and Arxiv are also included. In the above we summarizes the statistics of the collected papers. We see that the number of papers released about graph heterophily has significantly increased in recent five years. Meanwhile, the distribution of sources for collected papers published is given above. Further, we present the high-frequency keywords that appear in the titles of collected papers.

Surveys

  • [Arxiv 2024] A Survey on Learning from Graphs with Heterophily:Progress and Future [Paper]
  • [Arxiv 2024] The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challenges [Paper]
  • [IEEE DBE 2023] Heterophily and Graph Neural Networks: Past, Present and Future [Paper]
  • [Arxiv 2022] Graph Neural Networks for Graphs with Heterophily: A Survey [Paper]

Papers published per year

Year 2024

  • [NIPS 2024] On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks [Paper], [Code]

  • [IJCAI 2024] LG-GNN: Local-Global Adaptive Graph Neural Network for Modeling Both Homophily and Heterophily [Paper], [Code]

  • [IJCAI 2024] HeterGCL: Graph Contrastive Learning Framework on Heterophilic Graph [Paper], [Code]

  • [KDD 2024] Resurrecting Label Propagation for Graphs with Heterophily and Label Noise [Paper], [Code]

  • [KDD 2024] Unveiling Privacy Vulnerabilities: Investigating the Role of Structure in Graph Data [Paper], [Code]

  • [KDD 2024] AGS-GNN: Attribute-guided Sampling for Graph Neural Networks [Paper], [Code]

  • [KDD 2024] PolyFormer: Scalable Node-wise Filters via Polynomial Graph Transformer [Paper], [Code]

  • [KDD 2024] The Heterophilic Snowflake Hypothesis:Training and Empowering GNNs for Heterophilic Graphs [Paper], [Code]

  • [KDD 2024] Flexible Graph Neural Diffusion with Latent Class Representation Learning [Paper], [Code]

  • [ICML 2024] Multi-Track Message Passing: Tackling Oversmoothing and Oversquashing in Graph Learning via Preventing Heterophily Mixing [Paper], [Code]

  • [ICML 2024] Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs [Paper], [Code]

  • [ICML 2024] SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter [Paper], [Code]

  • [ICML 2024] Cooperative Graph Neural Networks [Paper], [Code]

  • [ICML 2024] GATE: How to Keep Out Intrusive Neighbors [Paper], [Code]

  • [ICML 2024] Less is More: on the Over-Globalizing Problem in Graph Transformers [Paper], [Code]

  • [ICML 2024] How Universal Polynomial Bases Enhance Spectral Graph Neural Networks: Heterophily, Over-smoothing, and Over-squashing [Paper], [Code]

  • [ICML 2024] Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspective [Paper], [Code]

  • [ICML 2024] Efficient Contrastive Learning for Fast and Accurate Inference on Graphs [Paper], [Code]

  • [ICML 2024] Mitigating Oversmoothing Through Reverse Process of GNNs for Heterophilic Graphs [Paper], [Code]

  • [ICML 2024] S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning [Paper], [Code]

  • [ICML 2024] Neurodegenerative Brain Network Classification via Adaptive Diffusion with Temporal Regularization [Paper], [Code]

  • [ICML 2024] Understanding Heterophily for Graph Neural Networks [Paper], [Code]

  • [ICLR 2024] VCR-graphormer: A mini-batch graph transformer via virtual connections [Paper], [Code]

  • [ICLR 2024] Affinity-based homophily: can we measure homophily of a graph without using node labels? [Paper], [Code]

  • [ICLR 2024] Partitioning Message Passing for Graph Fraud Detection [Paper], [Code]

  • [ICLR 2024] PolyGCL: Graph Contrastive Learning via Learnable Spectral Polynomial Filters [Paper], [Code]

  • [ICLR 2024] Polynormer: Polynomial-Expressive Graph Transformer in Linear Time [Paper], [Code]

  • [ICLR 2024-Tiny] Affinity-based Homophily: Can we measure homophily of a graph without using node labels? [Paper], [Code]

  • [WWW 2024] Graph Contrastive Learning Reimagined: Exploring Universality [Paper], [Code]

  • [WWW 2024] Optimizing Polynomial Graph Filters: A Novel Adaptive Krylov Subspace Approach [Paper], [Code]

  • [WWW 2024] Disambiguated Node Classification with Graph Neural Networks [Paper], [Code]

  • [WWW 2024] Challenging Low Homophily in Social Recommendation [Paper], [Code]

  • [WWW 2024] GAUSS: GrAph-customized Universal Self-Supervised Learning [Paper], [Code]

  • [AAAI 2024] PC-Conv: Unifying Homophily and Heterophily with Two-fold Filtering [Paper], [Code]

  • [AAAI 2024] Kumaraswamy Wavelet for Heterophilic Scene Graph Generation [Paper], [Code]

  • [AAAI 2024] DGA-GNN: Dynamic Grouping Aggregation GNN for Fraud Detection [Paper], [Code]

  • [AAAI 2024] Graph Neural Networks with Soft Association between Topology and Attribute [Paper], [Code]

  • [AAAI 2024] A generalized neural diffusion framework on graphs [Paper], [Code]

  • [AAAI 2024] Feature Transportation Improves Graph Neural Networks [Paper], [Code]

  • [AAAI 2024] Revisiting Graph-Based Fraud Detection in Sight of Heterophily and Spectrum [Paper], [Code]

  • [AAAI 2024] Homophily-Related: Adaptive Hybrid Graph Filter for Multi-View Graph Clustering [Paper], [Code]

  • [AAAI 2024] Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks [Paper], [Code]

  • [ICDE 2024] GraphRARE: Reinforcement Learning Enhanced Graph Neural Network with Relative Entropy [Paper], [Code]

  • [SIGIR 2024] SIGformer: Sign-aware Graph Transformer for Recommendation [Paper], [Code]

  • [MM 2024] Joint Homophily and Heterophily Relational Knowledge Distillation for Efficient and Compact 3D Object Detection [Paper], [Code]

  • [CVPR 2024] Improving Graph Contrastive Learning via Adaptive Positive Sampling [Paper], [Code]

  • [ECML-PKDD 2024] Self-Pro: A Self-Prompt and Tuning Framework for Graph Neural Networks [Paper], [Code]

  • [ECML-PKDD 2024] Enhancing Graph Neural Networks with Structure-Based Prompt [Paper], [Code]

  • [UAI 2024] Graph Contrastive Learning under Heterophily via Graph Filters [Paper], [Code]

  • [CIKM 2024] HC-GST: Heterophily-aware Distribution Consistency based Graph Self-training [Paper], [Code]

  • [WSDM 2024] Gad-nr: Graph anomaly detection via neighborhood reconstruction [Paper], [Code]

  • [LoG 2024] On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks [Paper], [Code]

  • [ICPR 2024] Heterophily-aware Social Bot Detection with Supervised Contrastive Learning [Paper], [Code]

  • [ACML 2024] Unleashing the Power of High-pass Filtering in Continuous Graph Neural Networks [Paper], [Code]

  • [BigComp 2024] Enhancing Heterophilic Graph Neural Network Performance Through Label Propagation in K-Nearest Neighbor Graphs [Paper], [Code]

  • [APSI 2024] On the Heterophily of Program Graphs: A Case Study of Graph-based Type Inference [Paper], [Code]

  • [ICPC 2024] A Just-in-time Software Defect Localization Method based on Code Graph Representation [Paper], [Code]

  • [IJMLC 2024] A Data-centric graph neural network for node classification of heterophilic networks [Paper], [Code]

  • [PNAS] Homophily modulates double descent generalization in graph convolution networks [Paper], [Code]

  • [TKDE] Make Heterophily Graphs Better Fit GNN: A Graph Rewiring Approach [Paper], [Code]

  • [TKDE] When Transformer Meets Large Graphs: An Expressive and Efficient Two-View Architecture [Paper], [Code]

  • [TKDE] Prioritized Propagation in Graph Neural Networks [Paper], [Code]

  • [TKDE] SpikeGraphormer: A High-Performance Graph Transformer with Spiking Graph Attention [Paper], [Code]

  • [TPAMI] Node-oriented Spectral Filtering for Graph Neural Networks [Paper], [Code]

  • [TPAMI] Graph Regulation Network for Point Cloud Segmentation [Paper], [Code]

  • [TMLR] Single-Pass Contrastive Learning Can Work for Both Homophilic and Heterophilic Graph? [Paper], [Code]

  • [TNNLS] Redundancy Is Not What You Need: An Embedding Fusion Graph Auto-Encoder for Self-Supervised Graph Representation Learning [Paper], [Code]

  • [TNNLS] Boosting Pseudo-Labeling With Curriculum Self-Reflection for Attributed Graph Clustering [Paper], [Code]

  • [TMCCA] TA-Detector: A GNN-based Anomaly Detector via Trust Relationship [Paper], [Code]

  • [Neurocomputing] WSSGCN: Wide Sub-stage Graph Convolutional Networks [Paper], [Code]

  • [Neural Networks] DPGCL: Dual pass filtering based graph contrastive learning [Paper], [Code]

  • [Pattern Recognition Letters] Neighbors selective Graph Convolutional Network for homophily and heterophily [Paper], [Code]

  • [Pattern Recognition Letters] Neighbors selective Graph Convolutional Network for homophily and heterophily [Paper], [Code]

  • [Knowledge-Based Systems] Portable graph-based rumour detection against multi-modal heterophily [Paper], [Code]

  • [Knowledge-Based Systems] SLGCN: Structure-enhanced line graph convolutional network for predicting drug–disease associations [Paper], [Code]

  • [Expert Systems with Applications] Multi-View Discriminative Edge Heterophily Contrastive Learning Network for Graph Anomaly Detection [Paper], [Code]

  • [Expert Systems with Applications] KNN-GNN: A powerful graph neural network enhanced by aggregating K-nearest neighbors in common subspace [Paper], [Code]

  • [Knowledge and Information Systems] SimGCL: graph contrastive learning by finding homophily in heterophily [Paper], [Code]

  • [World Wide Web] VR-GNN: Variational Relation Vector Graph Neural Network for Modeling both Homophily and Heterophily [Paper], [Code]

  • [OpenReview] MPformer: Advancing Graph Modeling Through Heterophily Relationship-Based Position Encoding [Paper], [Code]

  • [OpenReview] PROSPECT: Learn MLPs Robust against Graph Adversarial Structure Attacks [Paper], [Code]

  • [OpenReview] Molecule Generation by Heterophilious Triple Flows [Paper], [Code]

  • [Arxiv] NTFormer: A Composite Node Tokenized Graph Transformer for Node Classification [Paper], [Code]

  • [Arxiv] Learning Personalized Scoping for Graph Neural Networks under Heterophily [Paper], [Code]

  • [Arxiv] Differentiable Cluster Graph Neural Network [Paper], [Code]

  • [Arxiv] Are Heterophily-Specific GNNs and Homophily Metrics Really Effective? Evaluation Pitfalls and New Benchmarks [Paper], [Code]

  • [Arxiv] Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification [Paper], [Code]

  • [Arxiv] What Is Missing In Homophily? Disentangling Graph Homophily For Graph Neural Networks [Paper], [Code]

  • [Arxiv] Leveraging Contrastive Learning for Enhanced Node Representations in Tokenized Graph Transformers [Paper], [Code]

  • [Arxiv] NTFormer: A Composite Node Tokenized Graph Transformer for Node Classification [Paper], [Code]

  • [Arxiv] Node-wise Filtering in Graph Neural Networks: A Mixture of Experts Approach [Paper], [Code]

  • [Arxiv] Heterophily-Aware Fair Recommendation using Graph Convolutional Networks [Paper], [Code]

  • [Arxiv] Self-Attention Empowered Graph Convolutional Network for Structure Learning and Node Embedding [Paper], [Code]

  • [Arxiv] Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspective [Paper], [Code]

  • [Arxiv] Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Network [Paper], [Code]

  • [Arxiv] Link Prediction under Heterophily: A Physics-Inspired Graph Neural Network Approach [Paper], [Code]

  • [Arxiv] Molecule Generation by Heterophilious Triple Flows [Paper], [Code]

  • [Arxiv] Provable Filter for Real-world Graph Clustering [Paper], [Code]

  • [Arxiv] HENCLER: Node Clustering in Heterophilous Graphs through Learned Asymmetric Similarity [Paper], [Code]

  • [Arxiv] UniGAP: A Universal and Adaptive Graph Upsampling Approach to Mitigate Over-Smoothing in Node Classification Tasks [Paper], [Code]

  • [Arxiv] Learning a Mini-batch Graph Transformer via Two-stage Interaction Augmentation [Paper], [Code]

  • [Arxiv] Graph Triple Attention Network: A Decoupled Perspective [Paper], [Code]

  • [Arxiv] Urban Region Pre-training and Prompting: A Graph-based Approach [Paper], [Code]

  • [Arxiv] Non-Homophilic Graph Pre-Training and Prompt Learning [Paper], [Code]

  • [Arxiv] Provable Filter for Real-world Graph Clustering [Paper], [Code]

  • [Arxiv] Universally Robust Graph Neural Networks by Preserving Neighbor Similarity [Paper], [Code]

  • [Arxiv] On provable privacy vulnerabilities of graph representations [Paper], [Code]

  • [Arxiv] Learning to Model Graph Structural Information on MLPs via Graph Structure Self-Contrasting [Paper], [Code]

  • [Arxiv] Generation is better than Modification: Combating High Class Homophily Variance in Graph Anomaly Detection [Paper], [Code]

  • [Arxiv] Heterophily-Aware Fair Recommendation using Graph Convolutional Networks [Paper], [Code]

  • [Arxiv] When Heterophily Meets Heterogeneous Graphs: Latent Graphs Guided Unsupervised Representation Learning [Paper], [Code]

  • [Arxiv] When Heterophily Meets Heterogeneity: New Graph Benchmarks and Effective Methods [Paper], [Code]

  • [Arxiv] Leveraging Invariant Principle for Heterophilic Graph Structure Distribution Shifts [Paper], [Code]

  • [Arxiv] HC-GST: Heterophily-aware Distribution Consistency based Graph Self-training [Paper], [Code]

  • [Arxiv] Unraveling the Impact of Heterophilic Structures on Graph Positive-Unlabeled Learning [Paper], [Code]

  • [Arxiv] Exploring the Potential of Large Language Models for Heterophilic Graphs [Paper], [Code]

Year 2023

  • [ICLR 2023] Difformer: Scalable (graph) transformers induced by energy constrained diffusion [Paper], [Code]

  • [ICLR 2023] Specformer: Spectral Graph Neural Networks Meet Transformers [Paper], [Code]

  • [ICLR 2023] Gradient Gating for Deep Multi-Rate Learning on Graphs [Paper], [Code]

  • [ICLR 2023] ACMP: Allen-Cahn Message Passing with Attractive and Repulsive Forces for Graph Neural Networks [Paper], [Code]

  • [ICLR 2023] Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing [Paper], [Code]

  • [ICLR 2023] A Critical Look at Evaluation of GNNs Under Heterophily: Are We Really Making Progress? [Paper], [Code]

  • [ICLR 2023] GReTo: Remedying dynamic graph topology-task discordance via target homophily [Paper], [Code]

  • [ICLR-W 2023] Projections of Model Spaces for Latent Graph Inference [Paper], [Code]

  • [NeurIPS 2023] GSLB: The Graph Structure Learning Benchmark [Paper], [Code]

  • [NeurIPS 2023] Simplifying and empowering transformers for large-graph representations [Paper], [Code]

  • [NeurIPS 2023] Architecture matters: Uncovering implicit mechanisms in graph contrastive learning [Paper], [Code]

  • [NeurIPS 2023] A fractional graph laplacian approach to oversmoothing [Paper], [Code]

  • [NeurIPS 2023] FiGURe: Simple and Efficient Unsupervised Node Representations with Filter Augmentations [Paper], [Code]

  • [NeurIPS 2023] Truncated Affinity Maximization: One-class Homophily Modeling for Graph Anomaly Detection [Paper], [Code]

  • [NeurIPS 2023] OpenGSL: A Comprehensive Benchmark for Graph Structure Learning [Paper], [Code]

  • [NeurIPS 2023] Characterizing Graph Datasets for Node Classification: Homophily-Heterophily Dichotomy and Beyond [Paper], [Code]

  • [NeurIPS 2023] LD2:Scalable heterophilous graph neural network with decoupled embedding [Paper], [Code]

  • [NeurIPS 2023] Simple and Asymmetric Graph Contrastive Learning without Augmentations [Paper], [Code]

  • [NeurIPS 2023] When do graph neural networks help with node classification? investigating the homophily principle on node distinguishability [Paper], [[Code]]

  • [NeurIPS 2023] Demystifying Structural Disparity in Graph Neural Networks: Can One Size Fit All? [Paper], [Code]

  • [NeurIPS-W 2023] Node Mutual Information: Enhancing Graph Neural Networks for Heterophily [Paper], [Code]

  • [ICML 2023] GREAD: Graph Neural Reaction-Diffusion Networks [Paper], [Code]

  • [ICML 2023] Characterizing Graph Datasets for Node Classification:Homophily–Heterophily Dichotomy and Beyond [Paper], [Code]

  • [ICML 2023] Finding the Missing-half: Graph Complementary Learning for Homophily-prone and Heterophily-prone Graphs [Paper], [Code]

  • [ICML 2023] Contrastive Learning Meets Homophily: Two Birds with One Stone [Paper], [Code]

  • [ICML 2023] Beyond Homophily: Reconstructing Structure for Graph-agnostic Clustering [Paper], [Code]

  • [ICML 2023] Towards Deep Attention in Graph Neural Networks: Problems and Remedies [Paper], [Code]

  • [ICML 2023] Half-Hop: A graph upsampling approach for slowing down message passing [Paper], [Code]

  • [ICML 2023] GOAT: A Global Transformer on Large-scale Graphs [Paper], [Code]

  • [ICML 2023] Graph neural networks with learnable and optimal polynomial bases [Paper], [Code]

  • [ICML 2023] Goat: A global transformer on large- scale graphs [Paper], [Code]

  • [ICML-W] Evolving Computation Graphs [Paper], [Code]

  • [KDD 2023] Spatial Heterophily Aware Graph Neural Networks [Paper], [Code]

  • [KDD 2023] Clenshaw Graph Neural Networks [Paper], [Code]

  • [KDD 2023] Node classification beyond homophily: Towards a general solution [Paper], [Code]

  • [KDD 2023] HomoGCL: Rethinking Homophily in Graph Contrastive Learning [Paper], [Code]

  • [KDD-W 2023] Examining the Effects of Degree Distribution and Homophily in Graph Learning Models [Paper], [Code]

  • [WWW 2023] Robust Mid-Pass Filtering Graph Convolutional Networks [Paper], [Code]

  • [WWW 2023] Rethinking Structural Encodings: Adaptive Graph Transformer for Node Classification Task [Paper], [Code]

  • [WWW 2023] Addressing Heterophily in Graph Anomaly Detection: A Perspective of Graph Spectrum [Paper], Code]

  • [WWW 2023] Homophily-oriented Heterogeneous Graph Rewiring [Paper], [Code]

  • [WWW 2023] Auto-HeG: Automated Graph Neural Network on Heterophilic Graphs [Paper], [Code]

  • [WWW 2023] Label Information Enhanced Fraud Detection against Low Homophily in Graphs [Paper], [Code]

  • [WWW 2023] SE-GSL: A General and Effective Graph Structure Learning Framework through Structural Entropy Optimization [Paper], [Code]

  • [WWW 2023] Homophily-oriented Heterogeneous Graph Rewiring [Paper], [Code]

  • [IJCAI 2023] Beyond Homophily: Robust Graph Anomaly Detection via Neural Sparsification [Paper], [Code]

  • [IJCAI 2023] LSGNN: Towards General Graph Neural Network in Node Classification by Local Similarity [Paper], [Code]

  • [IJCAI 2023] Graph Neural Convection-Diffusion with Heterophily [Paper], [Code]

  • [IJCAI 2023] Gapformer: Graph Transformer with Graph Pooling for Node Classifcation [Paper], [Code]

  • [AAAI 2023] Beyond Smoothing: Unsupervised Graph Representation Learning with Edge Heterophily Discriminating [Pdf], [Code]

  • [AAAI 2023] Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering [Paper], [Code]

  • [AAAI-W 2023] 2-hop Neighbor Class Similarity (2NCS): A graph structural metric indicative of graph neural network performance [Paper], [Code]

  • [CVPR 2023] From Node Interaction To Hop Interaction: New Effective and Scalable Graph Learning Paradigm [Paper], [Code]

  • [MM 2023] Multi-modal Social Bot Detection: Learning Homophilic and Heterophilic Connections Adaptively [Paper], [Code]

  • [ICME 2023] Improving the Homophily of Heterophilic Graphs for Semi-Supervised Node Classification [Paper], [Code]

  • [CIKM 2023] HOVER: Homophilic Oversampling via Edge Removal for Class-Imbalanced Bot Detection on Graphs [Paper], [Code]

  • [CIKM 2023] Self-supervised Learning and Graph Classification under Heterophily [Paper], [Code]

  • [CIKM 2023] Neighborhood Homophily-based Graph Convolutional Network [Paper], [Code]

  • [CIKM 2023] Homophily-enhanced Structure Learning for Graph Clustering [Paper], [Code]

  • [CIKM 2023] MUSE: Multi-View Contrastive Learning for Heterophilic Graphs [Paper], [Code], [Code]

  • [CIKM 2023] SplitGNN: Spectral Graph Neural Network for Fraud Detection against Heterophily [Paper], [Code]

  • [ECML-PKDD 2023] Semi-Supervised Social Bot Detection with Initial Residual Relation Attention Networks [Paper], [Code]

  • [ECML-PKDD 2023] Leveraging Free Labels to Power up Heterophilic Graph Learning in Weakly-Supervised Settings: An Empirical Study [Paper], [Code]

  • [ECML-PKDD 2023] Learning to Augment Graph Structure for both Homophily and Heterophily [Paper], [Code]

  • [WSDM 2023] Alleviating Structural Distribution Shift in Graph Anomaly Detection [Paper], [Code]

  • [LoG 2023] On Performance Discrepancies Across Local Homophily Levels in Graph Neural Networks [Paper], [Code]

  • [LoG 2023] Geodesic Distributions Reveal How Heterophily and Bottlenecks Limit the Expressive Power of Message Passing Neural Networks [Paper], [Code]

  • [ICDM-W 2023] Dynamic Relation-Attentive Graph Neural Networks for Fraud Detection [Paper], [Code]

  • [Neurocomputing] Signed attention based graph neural network for graphs with heterophily [Paper], [Code]

  • [IEEE Access] Improved Modeling and Generalization Capabilities of Graph Neural Networks With Legendre Polynomials [Paper], [Code]

  • [TKDD] Multi-View Graph Representation Learning Beyond Homophily [Paper], [Code]

  • [TNNLS] Exploiting Neighbor Effect: Conv-Agnostic GNNs Framework for Graphs with Heterophily [Paper], [Code]

  • [TNNLS] Homophily-enhanced self-supervision for graph structure learning: Insights and directions [Paper], [Code]

  • [TMLR] Understanding convolution on graphs via energies [Paper], [Code]

  • [Expert Systems with Applications] Contrastive graph clustering with adaptive filter [Paper], [Code]

  • [Information Sciences] Similarity-navigated graph neural networks for node classification [Paper], [Code]

  • [Information Sciences] Taming over-smoothing representation on heterophilic graphs [Paper], [Code]

  • [Applied Soft Computing] Imbalanced node classification with Graph Neural Networks: A unified approach leveraging homophily and label information [Paper], [Code]

  • [OpenReview] Shape-aware Graph Spectral Learning [Paper], [Code]

  • [OpenReview] Low-Rank Graph Neural Networks Inspired by the Weak-balance Theory in Social Networks [Paper], [Code]

  • [OpenReview] Node Classification Beyond Homophily: Towards a General Solution [Paper], [Code]

  • [OpenReview] ProtoGNN: Prototype-Assisted Message Passing Framework for Non-Homophilous Graphs [Paper], [Code]

  • [OpenReview] From ChebNet to ChebGibbsNet [Paper], [Code]

  • [OpenReview] Wide Graph Neural Network [Paper], [Code]

  • [OpenReview] SlenderGNN: Accurate, Robust, and Interpretable GNN, and the Reasons for its Success [Paper], [Code]

  • [OpenReview] ReD-GCN: Revisit the Depth of Graph Convolutional Network [Paper], [Code]

  • [OpenReview] Graph Neural Networks as Gradient Flows: Understanding Graph Convolutions via Energy [Paper], [Code]

  • [OpenReview] Are Graph Attention Networks Attentive Enough? Rethinking Graph Attention by Capturing Homophily and Heterophily [Paper], [Code]

  • [OpenReview] Causally-guided Regularization of Graph Attention improves Generalizability [Paper], [Code]

  • [Arxiv] Unifying over-smoothing and over-squashing in graph neural networks: A physics informed approach and beyond [Paper], [Code]

  • [Arxiv] From Continuous Dynamics to Graph Neural Networks: Neural Diffusion and Beyond [Paper], [Code]

  • [Arxiv] p-Laplacian Transformer [Paper], [Code]

  • [Arxiv] NP^2L: Negative Pseudo Partial Labels Extraction for Graph Neural Networks [Paper], [Code]

  • [Arxiv] SignGT: Signed Attention-based Graph Transformer for Graph Representation Learning [Paper], [Code]

  • [Arxiv] Contrastive Learning for Non-Local Graphs with Multi-Resolution Structural Views [Paper], [Code]

  • [Arxiv] Domain-adaptive Message Passing Graph Neural Network [Paper], [Code]

  • [Arxiv] Extended Graph Assessment Metrics for Graph Neural Networks [Paper], [Code]

  • [Arxiv] Muti-scale Graph Neural Network with Signed-attention for Social Bot Detection: A Frequency Perspective [Paper], [Code]

  • [Arxiv] Automated Polynomial Filter Learning for Graph Neural Networks [Paper], [Code]

  • [Arxiv] Frameless Graph Knowledge Distillation [Paper], [Code]

  • [Arxiv] QDC: Quantum Diffusion Convolution Kernels on Graphs [Paper], [Code]

  • [Arxiv] Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters [Paper], [Code]

  • [Arxiv] HOFA: Twitter Bot Detection with Homophily-Oriented Augmentation and Frequency Adaptive Attention [Paper], [Code]

  • [Arxiv] Self-supervised Learning and Graph Classification under Heterophily [Paper], [Code]

  • ArXiv] GPatcher: A Simple and Adaptive MLP Model for Alleviating Graph Heterophily [Paper], [Code]

  • [Arxiv] PathMLP: Smooth Path Towards High-order Homophily [Paper], [Code]

  • [Arxiv] Permutation Equivariant Graph Framelets for Heterophilous Graph Learning [Paper], [Code]

  • [Arxiv] Edge Directionality Improves Learning on Heterophilic Graphs [Paper], [Code]

  • [Arxiv] Addressing Heterophily in Node Classification with Graph Echo State Networks [Paper], [Code]

  • [Arxiv] A Fractional Graph Laplacian Approach to Oversmoothing [Paper], [Code]

  • [Arxiv] From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module [Paper], [Code]

  • [Arxiv] Self-attention Dual Embedding for Graphs with Heterophily [Paper], [Code]

  • [Arxiv] SIMGA: A Simple and Effective Heterophilous Graph Neural Network with Efficient Global Aggregation [Paper], [Code]

  • [Arxiv] Imbalanced Node Classification Beyond Homophilic Assumption [Paper], [Code]

  • [Arxiv] Graph Positional Encoding via Random Feature Propagation [Paper], [Code]

  • [Arxiv] Steering Graph Neural Networks with Pinning Control [Paper], [Code]

  • [Arxiv] Attending to Graph Transformers Paper], [Code]

  • [Arxiv] Heterophily-Aware Graph Attention Network [Paper], [Code]

  • [Arxiv] Semi-Supervised Classification with Graph Convolutional Kernel Machines [Paper], [Code]

  • [Arxiv] A Graph Neural Network with Negative Message Passing for Graph Coloring [Paper], [Code]

  • [Arxiv] Is Signed Message Essential for Graph Neural Networks? [Paper], [Code]

  • [Arxiv] Self-attention Dual Embedding for Graphs with Heterophily [Paper], [Code]

  • [Arxiv] The Snowflake Hypothesis: Training Deep GNN with One Node One Receptive field [Paper], [Code]

  • [Arxiv] Contrastive Learning for Non-Local Graphs with Multi-Resolution Structural Views [Paper], [Code]

  • [Arxiv] Privacy-Utility Trade-offs in Neural Networks for Medical Population Graphs: Insights from Differential Privacy and Graph Structure [Paper], [Code]

  • [Arxiv] Muti-scale Graph Neural Network with Signed-attention for Social Bot Detection: A Frequency Perspective [Paper], [Code]

  • [Arxiv] Hetero2Net: Heterophily-aware Representation Learning on Heterogenerous Graphs [Paper], [Code]

Year 2022

  • [ICLR 2022] Graph Auto-Encoder Via Neighborhood Wasserstein Reconstruction [Paper], [Code]

  • [ICLR 2022] Understanding over-squashing and bottlenecks on graphs via curvature [Paper], [Code]

  • [ICLR 2022] Is Homophily A Necessity for Graph Neural Networks? [Paper], [Code]

  • [ICLR 2022] Understanding and Improving Graph Injection Attack by Promoting Unnoticeability [Paper], [Code]

  • [ICLR 2022] Neural Link Prediction with Walk Pooling [Paper], [Code]

  • [NeurIPS 2022] Hierarchical Graph Transformer with Adaptive Node Sampling [Paper], [Code]

  • [NeurIPS 2022] EvenNet: Ignoring Odd-Hop Neighbors Improves Robustness of Graph Neural Networks [Paper], [Code]

  • [NeurIPS 2022] Decoupled Self-supervised Learning for Non-Homophilous Graphs [Paper], [Code]

  • [NeurIPS 2022] NodeFormer: A Scalable Graph Structure Learning Transformer for Node Classification [Paper], [Code]

  • [NeurIPS 2022] Simplified Graph Convolution with Heterophily [Paper], [Code]

  • [NeurIPS 2022] Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs [Paper], [Code]

  • [NeurIPS 2022] Revisiting Heterophily For Graph Neural Networks [Paper], [Code]

  • [NeurIPS 2022] Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited [Paper], [Code]

  • [NeurIPS-W 2022] From Local to Global: Spectral-Inspired Graph Neural Networks [Paper], [Code]

  • [NeurIPS-W 2022] Complete the Missing Half: Augmenting Aggregation Filtering with Diversification for Graph Convolutional Networks [Paper], [Code]

  • [ICML 2022] G2CN: Graph gaussian convolution networks with concentrated graph filters [Paper], [Code]

  • [ICML 2022] Finding Global Homophily in Graph Neural Networks When Meeting Heterophily [Paper], [Code]

  • [ICML 2022] How Powerful are Spectral Graph Neural Networks [Paper], [Code]

  • [ICML 2022] Rethinking graph neural networks for anomaly detection [Paper], [Code]

  • [ICML 2022] Optimization-Induced Graph Implicit Nonlinear Diffusion [Paper], [Code]

  • [ICML-W 2022] Sheaf Neural Networks with Connection Laplacians [Paper], [Code]

  • [KDD 2022] How does Heterophily Impact Robustness of Graph Neural Networks? Theoretical Connections and Practical Implications [Paper], [Code]

  • [KDD-W 2022] On Graph Neural Network Fairness in the Presence of Heterophilous Neighborhoods [Paper], [Code]

  • [WWW 2022] GBK-GNN: Gated Bi-Kernel Graph Neural Networks for Modeling Both Homophily and Heterophily [Paper], [Code]

  • [WWW 2022] H2-FDetector: A GNN-based Fraud Detector with Homophilic and Heterophilic Connections [Paper], [Code]

  • [IJCAI 2022] Beyond Homophily: Structure-aware Path Aggregation Graph Neural Network [Paper], [Code]

  • [IJCAI 2022] RAW-GNN: RAndom Walk Aggregation based Graph Neural Network [Paper], [Code]

  • [AAAI 2022] Deformable Graph Transformer [Paper], [Code]

  • [AAAI 2022] Block Modeling-Guided Graph Convolutional Neural Networks [Paper], [Code]

  • [AAAI 2022] Powerful Graph Convolutioal Networks with Adaptive Propagation Mechanism for Homophily and Heterophily [Paper], [Code]

  • [AAAI 2022] Deformable Graph Convolutional Networks [Paper], [Code]

  • [AAAI 2022] Graph Pointer Neural Networks [Paper], [Code]

  • [AAAI 2022] RAW-GNN: Random walk aggregation based graph neural network [Paper], [Code]

  • [CVPR 2022] HL-Net: Heterophily Learning Network for Scene Graph Generation [Paper], [Code]

  • [CIKM 2022] Finding heterophilic neighbors via confidence-based subgraph matching for semi-supervised node classification [Paper], [Code]

  • [CIKM 2022] Towards Self-supervised Learning on Graphs with Heterophily [Paper], [Code]

  • [ICASSP 2022] Memory-based Message Passing: Decoupling the Message for Propogation from Discrimination [Paper], [Code]

  • [LoG 2022] Label-Wise Message Passing Graph Neural Network on Heterophilic Graphs [Paper], [Code]

  • [LoG 2022] Leave Graphs Alone: Addressing Over-Squashing without Rewiring [Paper], [Code]

  • [LoG 2022] Global-Local Graph Neural Networks for Node-Classification [Paper], [Code]

  • [LoG 2022] GARNET: Reduced-Rank Topology Learning for Robust and Scalable Graph Neural Networks [Paper], [Code]

  • [LoG 2022] DiffWire: Inductive Graph Rewiring via the Lovász Bound [Paper], [Code]

  • [LoG 2022] [Tutorial] Graph Rewiring: From Theory to Applications in Fairness [Link], [Code]

  • [ICDM 2022] HP-GMN: Graph Memory Networks for Heterophilous Graphs [Paper], [Code]

  • [ICDM 2022] Two Sides of The Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks [Paper], [Code]

  • [ICDM-W 2022] Improving Your Graph Neural Networks: A High-Frequency Booster [Paper], [Code]

  • [TMLR] Unsupervised Network Embedding Beyond Homophily [Paper], [Code]

  • [TNNLS] NCGNN: Node-Level Capsule Graph Neural Network for Semisupervised Classification [Paper], [Code]

  • [TKDE] Beyond Low-pass Filtering: Graph Convolutional Networks with Automatic Filtering [Paper], [Code]

  • [Knowledge-Based Systems] Multi-view learning with distinguishable feature fusion for rumor detection [Paper], [Code]

  • [Arxiv] Revisiting Heterophily in Graph Convolution Networks by Learning Representations Across Topological and Feature Spaces [Paper], [Code]

  • [Arxiv] GLINKX: A Scalable Unified Framework For Homophilous and Heterophilous Graphs [Paper], [Code]

  • [Arxiv] Unifying Label-inputted Graph Neural Networks with Deep Equilibrium Models [Paper], [Code]

  • [Arxiv] Neighborhood Convolutional Network: A New Paradigm of Graph Neural Networks for Node Classification [Paper], [Code]

  • [Arxiv] Enhancing Intra-class Information Extraction for Heterophilous Graphs: One Neural Architecture Search Approach [Paper], [Code]

  • [Arxiv] Transductive Kernels for Gaussian Processes on Graphs [Paper], [Code]

  • [Arxiv] Flip Initial Features: Generalization of Neural Networks for Semi-supervised Node Classification [Paper], [Code]

  • [Arxiv] When Do We Need GNN for Node Classification? [Paper], [Code]

  • [Arxiv] Break the Wall Between Homophily and Heterophily for Graph Representation Learning [Paper], [Code]

  • [Arxiv] GPNet: Simplifying Graph Neural Networks via Multi-channel Geometric Polynomials [Paper], [Code]

  • [Arxiv] Graph Polynomial Convolution Models for Node Classification of Non-Homophilous Graphs [Paper], [Code]

  • [Arxiv] Link Prediction on Heterophilic Graphs via Disentangled Representation Learning [Paper], [Code]

  • [Arxiv] What Do Graph Convolutional Neural Networks Learn? [Paper], [Code]

  • [Arxiv] Demystifying Graph Convolution with a Simple Concatenation [Paper], [Code]

  • [Arxiv] Restructuring Graph for Higher Homophily via Learnable Spectral Clustering [Paper], [Code]

  • [Arxiv] Graph Neural Networks as Gradient Flows [Paper], [Code]

  • [Arxiv] Augmentation-Free Graph Contrastive Learning with Performance Guarantee [Paper], [Code]

  • [Arxiv] Simplifying Node Classification on Heterophilous Graphs with Compatible Label Propagation [Paper], [Code]

  • [Arxiv] Revisiting the Role of Heterophily in Graph Representation Learning: An Edge Classification Perspective [Paper], [Code]

  • [Arxiv] Graph Representation Learning Beyond Node and Homophily [Paper], [Code]

  • [Arxiv] Incorporating Heterophily into Graph Neural Networks for Graph Classification [Paper], [Code]

  • [Arxiv] GSN: A Universal Graph Neural Network Inspired by Spring Network [Paper], [Code]

  • [Arxiv] Enhancing Intra-class Information Extraction for Heterophilous Graphs: One Neural Architecture Search Approach [Paper], [Code]

Year 2021

  • [ICLR 2021] Adaptive Universal Generalized PageRank Graph Neural Network [Paper], [Code]

  • [ICLR 2021] How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision [Paper], [Code]

  • [NeurIPS 2021] Pde-gcn: Novel architectures for graph neural networks motivated by partial differential equations [Paper], [Code]

  • [NeurIPS 2021] Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods [Paper], [Code]

  • [NeurIPS 2021] Diverse Message Passing for Attribute with Heterophily [Paper], [Code]

  • [NeurIPS 2021] Universal Graph Convolutional Networks [Paper], [Code]

  • [NeurIPS 2021] EIGNN: Efficient Infinite-Depth Graph Neural Networks [Paper], [Code]

  • [NeurIPS 2021] BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation [Paper], [Code]

  • [KDD 2021] Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns [Paper], [Code]

  • [WWW 2021] New Benchmarks for Learning on Non-Homophilous Graphs [Paper], [Code]

  • [WWW 2021] Graph Structure Estimation Neural Networks [Paper], [Code]

  • [AAAI 2021] Beyond Low-frequency Information in Graph Convolutional Networks [Paper], [Code]

  • [AAAI 2021] Graph Neural Networks with Heterophily [Paper], [Code]

  • [CIKM 2021] AdaGNN: Graph Neural Networks with Adaptive Frequency Response Filter [Paper], [Code]

  • [CIKM 2021] Tree Decomposed Graph Neural Network [Paper], [Code]

  • [WSDM 2021] Node Similarity Preserving Graph Convolutional Networks [Paper], [Code]

  • [ICASSP] Geometric Scattering Attention Networks [Paper], [Code]

  • [TPAMI] Non-Local Graph Neural Networks [Paper], [Code]

  • [Journal of Physics: Conference Series] Energy Levels Based Graph Neural Networks for Heterophily [Paper], [Code]

  • [Arxiv] Unifying Homophily and Heterophily Network Transformation via Motifs [Paper], [Code]

  • [Arxiv] SkipNode: On Alleviating Over-smoothing for Deep Graph Convolutional Networks [Paper], [Code]

  • [Arxiv] Simplifying Approach to Node Classification in Graph Neural Networks [Paper], [Code]

  • [Arxiv] GCN-SL: Graph Convolutional Networks with Structure Learning for Graphs under Heterophily [Paper], [Code]

  • [Arxiv] Beyond Low-Pass Filters: Adaptive Feature Propagation on Graphs [Paper], [Code]

  • [Arxiv] Node2Seq: Towards Trainable Convolutions in Graph Neural Networks [Paper], [Code]

  • [Arxiv] Graph Decoupling Attention Markov Networks for Semi-supervised Graph Node Classification [Paper], [Code]

  • [Arxiv] Relational Graph Neural Network Design via Progressive Neural Architecture Search [Paper], [Code]

Year 2020

  • [ICLR 2020] Geom-GCN: Geometric Graph Convolutional Networks [Paper], [Code]

  • [ICML 2020] Simple and Deep Graph Convolutional Networks [Paper] [Code]

  • [NeurIPS 2020] Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs [Paper], [Code]

Year 2019

  • [ICML 2019] MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing [Paper], [Code]