README.md

Publications in ICLR 2024

Scalability

• Scalable and Effective Implicit Graph Neural Networks on Large Graphs
• Scalable Modular Network: A Framework for Adaptive Learning via Agreement Routing
• Forward Learning of Graph Neural Networks
• LogicMP: A Neuro-symbolic Approach for Encoding First-order Logic Constraints
• DREAM: Dual Structured Exploration with Mixup for Open-set Graph Domain Adaption
• CoRe-GD: A Hierarchical Framework for Scalable Graph Visualization with GNNs

Latent Graph Inference

• Neural Snowflakes: Universal Latent Graph Inference via Trainable Latent Geometries
• From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module

Oversmoothing

• Graph Parsing Networks
• Boosting Graph Anomaly Detection with Adaptive Message Passing

Long-range Dependencies

• Long-Short-Range Message-Passing: A Fragmentation-Based Framework to Capture Non-Local Atomistic Interactions
• Unleashing the Potential of Fractional Calculus in Graph Neural Networks with FROND
• Locality-Aware Graph Rewiring in GNNs

Graph Transformers

• Efficient Sharpness-Aware Minimization for Molecular Graph Transformer Models
• Transformers vs. Message Passing GNNs: Distinguished in Uniform
• Learning Adaptive Multiresolution Transforms via Meta-Framelet-based Graph Convolutional Network
• VCR-Graphormer: A Mini-batch Graph Transformer via Virtual Connections

Generalisability

• InterpGNN: Understand and Improve Generalization Ability of Transdutive GNNs through the Lens of Interplay between Train and Test Nodes
• The Optimal Constant Solution: Predictable Extrapolation in Deep Neural Networks
• Online GNN Evaluation Under Test-time Graph Distribution Shifts

Expressivity

• Beyond Weisfeiler-Lehman: A Quantitative Framework for GNN Expressiveness
• Rethinking and Extending the Probabilistic Inference Capacity of GNNs
• Learning From Simplicial Data Based on Random Walks and 1D Convolutions
• G$\textasciicircum2$N$\textasciicircum2$ : Weisfeiler and Lehman go grammatical
• On the Power of the Weisfeiler-Leman Test for Graph Motif Parameters
• Polynormer: Polynomial-Expressive Graph Transformer in Linear Time
• Efficient Subgraph GNNs by Learning Effective Selection Policies
• Effective Structural Encodings via Local Curvature Profiles
• Understanding Expressivity of Neural KG Reasoning from Rule Structure Learning
• Probabilistically Rewired Message-Passing Neural Networks
• Counting Graph Substructures with Graph Neural Networks
• On the Stability of Expressive Positional Encodings for Graph Neural Networks
• Rethinking the Power of Graph Canonization in Graph Representation Learning with Stability

Explainability

• Entropy Coding of Unordered Data Structures
• GraphChef: Decision-Tree Recipes to Explain Graph Neural Networks
• Learning to solve Class-Constrained Bin Packing Problems via Encoder-Decoder Model
• Graph Lottery Ticket Automated
• GNNBoundary: Towards Explaining Graph Neural Networks through the Lens of Decision Boundaries
• Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks
• REFACTOR: Learning to Extract Theorems from Proofs
• GOAt: Explaining Graph Neural Networks via Graph Output Attribution
• UNR-Explainer: Counterfactual Explanations for Unsupervised Node Representation Learning Models
• Conformal Inductive Graph Neural Networks
• Mixture of Weak and Strong Experts on Graphs
• Matrix Manifold Neural Networks++
• GnnX-Bench: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking
• Rayleigh Quotient Graph Neural Networks for Graph-level Anomaly Detection

Fairness

• Deceptive Fairness Attacks on Graphs via Meta Learning
• Aligning Relational Learning with Lipschitz Fairness
• TASK PLANNING FOR VISUAL ROOM REARRANGEMENT UNDER PARTIAL OBSERVABILITY
• Structural Fairness-aware Active Learning for Graph Neural Networks
• Pose Modulated Avatars from Video
• Adversarial Attacks on Fairness of Graph Neural Networks

Federated Learning

• FedWon: Triumphing Multi-domain Federated Learning Without Normalization
• VFLAIR: A Research Library and Benchmark for Vertical Federated Learning

Temporal Data

• Causality-Inspired Spatial-Temporal Explanations for Dynamic Graph Neural Networks
• Deep Temporal Graph Clustering
• VBH-GNN: Variational Bayesian Heterogeneous Graph Neural Networks for Cross-subject Emotion Recognition
• Graph Transformers on EHRs: Better Representation Improves Downstream Performance
• A Generative Pre-Training Framework for Spatio-Temporal Graph Transfer Learning
• Temporal Generalization Estimation in Evolving Graphs
• Graphpulse: Topological representations for temporal graph property prediction
• PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks
• NP-GL: Extending Power of Nature from Binary Problems to Real-World Graph Learning
• Decoding Natural Images from EEG for Object Recognition
• Graph-based Virtual Sensing from Sparse and Partial Multivariate Observations

Spatio-Temporal Data

• Beyond Spatio-Temporal Representations: Evolving Fourier Transform for Temporal Graphs
• TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts

Time Series

• Biased Temporal Convolution Graph Network for Time Series Forecasting with Missing Values.
• Amortized Network Intervention to Steer the Excitatory Point Processes

Spectral Methods

• HoloNets: Spectral Convolutions do extend to Directed Graphs
• PolyGCL: GRAPH CONTRASTIVE LEARNING via Learnable Spectral Polynomial Filters

Self-Supervision

• Self-supervised Heterogeneous Graph Learning: a Homogeneity and Heterogeneity Perspective
• HypeBoy: Generative Self-Supervised Representation Learning on Hypergraphs
• $\textbackslashmathbb\D\textasciicircum2$ Pruning: Message Passing for Balancing Diversity & Difficulty in Data Pruning

Contrastive Learning

• Multimodal Patient Representation Learning with Missing Modalities and Labels
• StructComp: Substituting propagation with Structural Compression in Training Graph Contrastive Learning
• A Graph is Worth 1-bit Spikes: When Graph Contrastive Learning Meets Spiking Neural Networks

Data Augmentation

• Consistency Training with Learnable Data Augmentation for Graph Anomaly Detection with Limited Supervision
• SAFLEX: Self-Adaptive Augmentation via Feature Label Extrapolation

Graph Generation

• HiGen: Hierarchical Graph Generative Networks
• Graph Generation with $K\textasciicircum2$-trees
• A Simple and Scalable Representation for Graph Generation

Molecules

• Learning Scalar Fields for Molecular Docking with Fast Fourier Transforms
• GTMGC: Using Graph Transformer to Predict Molecule\textquoterights Ground-State Conformation
• Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks
• Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs
• Protein-Ligand Interaction Prior for Binding-aware 3D Molecule Diffusion Models
• Stochastic Gradient Descent for Gaussian Processes Done Right
• Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets
• Removing Biases from Molecular Representations via Information Maximization
• From Molecules to Materials: Pre-training Large Generalizable Models for Atomic Property Prediction
• Navigating the Design Space of Equivariant Diffusion-Based Generative Models for De Novo 3D Molecule Generation

Proteins

• Learning to design protein-protein interactions with enhanced generalization
• Evaluating Representation Learning on the Protein Structure Universe
• Generative Adversarial Policy Network for Modelling Protein Complexes
• KW-Design: Pushing the Limit of Protein Deign via Knowledge Refinement
• Self-supervised Pocket Pretraining via Protein Fragment-Surroundings Alignment
• De novo Protein Design Using Geometric Vector Field Networks
• Rigid Protein-Protein Docking via Equivariant Elliptic-Paraboloid Interface Prediction
• Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design

Atoms

• Crystalformer: Infinitely Connected Attention for Periodic Structure Encoding
• Long-range Neural Atom Learning for Molecular Graphs
• Complete and Efficient Graph Transformers for Crystal Material Property Prediction

Equivariance

• Graph Neural Networks for Learning Equivariant Representations of Neural Networks
• On the hardness of learning under symmetries
• Orbit-Equivariant Graph Neural Networks
• Fast, Expressive $\textbackslashmathrm\SE(n)$ Equivariant Networks through Weight-Sharing in Position-Orientation Space
• Improving Generalization in Equivariant Graph Neural Networks with Physical Inductive Biases
• A Characterization Theorem for Equivariant Networks with Point-wise Activations
• EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations
• SaNN: Simple Yet Powerful Simplicial-aware Neural Networks
• Equivariant Matrix Function Neural Networks
• Hybrid Directional Graph Neural Network for Molecules
• Clifford Group Equivariant Simplicial Message Passing Networks
• Graph Metanetworks for Processing Diverse Neural Architectures
• Rethinking the Benefits of Steerable Features in 3D Equivariant Graph Neural Networks
• Energy-conserving equivariant GNN for elasticity of lattice architected metamaterials

Robustness

• Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attacks
• MMD Graph Kernel: Effective Metric Learning for Graphs via Maximum Mean Discrepancy
• GraphGuard: Provably Robust Graph Classification against Adversarial Attacks
• Mitigating Severe Robustness Degradation on Graphs
• Rethinking Label Poisoning for GNNs: Pitfalls and Attacks
• iGraphMix: Input Graph Mixup Method for Node Classification
• Robust Angular Synchronization via Directed Graph Neural Networks
• Boosting the Adversarial Robustness of Graph Neural Networks: An OOD Perspective

Knowledge Graphs

• Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors
• Towards Foundation Models for Knowledge Graph Reasoning
• GraphCare: Enhancing Healthcare Predictions with Personalized Knowledge Graphs

Large Language Models

• Label-free Node Classification on Graphs with Large Language Models (LLMs)
• Harnessing Explanations: LLM-to-LM Interpreter for Enhanced Text-Attributed Graph Representation Learning

Diffusion

• InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior
• MOFDiff: Coarse-grained Diffusion for Metal-Organic Framework Design
• Local Graph Clustering with Noisy Labels
• Hypergraph Dynamic System

Physics

• Space and time continuous physics simulation from partial observations
• Learning 3D Particle-based Simulators from RGB-D Videos
• AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction
• Symmetric Basis Convolutions for Learning Lagrangian Fluid Mechanics
• BroGNet: Momentum-Conserving Graph Neural Stochastic Differential Equation for Learning Brownian Dynamics
• Towards Cross Domain Generalization of Hamiltonian Representation via Meta Learning

Pre-Training

• Q-TAPE: A Task-Agnostic Pre-Trained Approach for Quantum Properties Estimation
• Decoupling Weighing and Selecting for Integrating Multiple Graph Pre-training Tasks

Distillation

• Training Graph Transformers via Curriculum-Enhanced Attention Distillation
• Mirage: Model-agnostic Graph Distillation for Graph Classification
• VQGraph: Rethinking Graph Representation Space for Bridging GNNs and MLPs
• TEDDY: Trimming Edges with Degree-based Graph Diffusion Strategy
• LightHGNN: Distilling Hypergraph Neural Networks into MLPs for 100x Faster Inference

Reinforcement Learning

• Deep Reinforcement Learning Guided Improvement Heuristic for Job Shop Scheduling
• Learning Multi-Agent Communication from Graph Modeling Perspective
• Towards Imitation Learning to Branch for MIP: A Hybrid Reinforcement Learning based Sample Augmentation Approach

Uncertainty

• A Stochastic Centering Framework for Improving Calibration in Graph Neural Networks
• Uncertainty-aware Graph-based Hyperspectral Image Classification

Link Prediction

• ETGraph: A Pioneering Dataset Bridging Ethereum and Twitter
• Neural Common Neighbor with Completion for Link Prediction
• NetInfoF Framework: Measuring and Exploiting Network Usable Information
• Revisiting Link Prediction: a data perspective
• A Topological Perspective on Demystifying GNN-Based Link Prediction Performance

Meshes

• Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformer
• Better Neural PDE Solvers Through Data-Free Mesh Movers

Miscellaneous

• Combining Axes Preconditioners through Kronecker Approximation for Deep Learning
• One For All: Towards Training One Graph Model For All Classification Tasks
• PROGRAM: PROtotype GRAph Model based Pseudo-Label Learning for Test-Time Adaptation
• Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks
• Simplicial Representation Learning with Neural $k$-Forms
• Variance-enlarged Poisson Learning for Graph-based Semi-Supervised Learning with Extremely Sparse Labeled Data
• NeuroBack: Improving CDCL SAT Solving using Graph Neural Networks
• Partitioning Message Passing for Graph Fraud Detection
• Encoding Unitig-level Assembly Graphs with Heterophilous Constraints for Metagenomic Contigs Binning
• BENO: Boundary-embedded Neural Operators for Elliptic PDEs