OGB
ogbg-molhiv
Structure
41,127Graphs
26Nodes
28Edges
2Classes
9Node Feature Dim
3Edge Feature Dim
Coverage
369Results
1Variants
73Papers
- ICML 2026: RADE: Random Add-Drop Edge as a Regularizer RADE: Random Add-Drop Edge as a Regularizer ICML 2026
- SigGate-GT: Taming Over-Smoothing in Graph Transformers via Sigmoid-Gated Attention SigGate-GT: Taming Over-Smoothing in Graph Transformers via Sigmoid-Gated Attention
- Multi-RF Fusion with Multi-GNN Blending for Molecular Property Prediction Multi-RF Fusion with Multi-GNN Blending for Molecular Property Prediction
- ICML 2025: GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning ICML 2025
- Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks
- ICML 2025: Positional Encoding meets Persistent Homology on Graphs Positional Encoding meets Persistent Homology on Graphs ICML 2025
- TMLR 2025: Flow-Attentional Graph Neural Networks Flow-Attentional Graph Neural Networks TMLR 2025
- Convexified Message-Passing Graph Neural Networks Convexified Message-Passing Graph Neural Networks
- WWW 2025: Graph Self-Supervised Learning with Learnable Structural and Positional Encodings Graph Self-Supervised Learning with Learnable Structural and Positional Encodings WWW 2025
- ICML 2025: Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks ICML 2025
- TMLR 2025: Generative Risk Minimization for Out-of-Distribution Generalization on Graphs Generative Risk Minimization for Out-of-Distribution Generalization on Graphs TMLR 2025
- Enhancing Distribution and Label Consistency for Graph Out-of-Distribution Generalization Enhancing Distribution and Label Consistency for Graph Out-of-Distribution Generalization
- A Hierarchical Language Model For Interpretable Graph Reasoning A Hierarchical Language Model For Interpretable Graph Reasoning
- Towards Few-shot Self-explaining Graph Neural Networks Towards Few-shot Self-explaining Graph Neural Networks
- TMLR 2024: Graph Pooling via Ricci Flow Graph Pooling via Ricci Flow TMLR 2024
- NeurIPS 2024: DiGRAF: Diffeomorphic Graph-Adaptive Activation Function DiGRAF: Diffeomorphic Graph-Adaptive Activation Function NeurIPS 2024
- AAAI 2024: A Scalable and Effective Alternative to Graph Transformers A Scalable and Effective Alternative to Graph Transformers AAAI 2024
- Scalable Expressiveness through Preprocessed Graph Perturbations Scalable Expressiveness through Preprocessed Graph Perturbations
- ICML 2024: How Interpretable Are Interpretable Graph Neural Networks? How Interpretable Are Interpretable Graph Neural Networks? ICML 2024
- Automatic Graph Topology-Aware Transformer Automatic Graph Topology-Aware Transformer
- NeurIPS 2024: GRANOLA: Adaptive Normalization for Graph Neural Networks GRANOLA: Adaptive Normalization for Graph Neural Networks NeurIPS 2024
- RandAlign: A Parameter-Free Method for Regularizing Graph Convolutional Networks RandAlign: A Parameter-Free Method for Regularizing Graph Convolutional Networks
- TMLR 2024: Contextualized Messages Boost Graph Representations Contextualized Messages Boost Graph Representations TMLR 2024
- Graph Learning with Distributional Edge Layouts Graph Learning with Distributional Edge Layouts
- Two Trades are not Baffled: Condensing Graph via Crafting Rational Gradient Matching Two Trades are not Baffled: Condensing Graph via Crafting Rational Gradient Matching
- ICLR 2024: PF-GNN: Differentiable particle filtering based approximation of universal graph representations PF-GNN: Differentiable particle filtering based approximation of universal graph representations ICLR 2024
- Permutation-Invariant graph partitioning: How graph neural networks capture structural interactions? Permutation-Invariant graph partitioning: How graph neural networks capture structural interactions?
- ICLR 2023: Efficient Subgraph GNNs by Learning Effective Selection Policies Efficient Subgraph GNNs by Learning Effective Selection Policies ICLR 2023
- NeurIPS 2023: MAG-GNN: Reinforcement Learning Boosted Graph Neural Network MAG-GNN: Reinforcement Learning Boosted Graph Neural Network NeurIPS 2023
- BiKT: Unleashing the potential of GNNs via Bi-directional Knowledge Transfer BiKT: Unleashing the potential of GNNs via Bi-directional Knowledge Transfer
- ICLR 2023: GnnX-Bench: Unravelling the Utility of Perturbation-based Gnn Explainers through In-depth Benchmarking GnnX-Bench: Unravelling the Utility of Perturbation-based Gnn Explainers through In-depth Benchmarking ICLR 2023
- Deep Prompt Tuning for Graph Transformers Deep Prompt Tuning for Graph Transformers
- Rethinking the Power of Graph Canonization in Graph Representation Learning with Stability Rethinking the Power of Graph Canonization in Graph Representation Learning with Stability
- AAAI 2023: Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes AAAI 2023
- NeurIPS 2023: Fused Gromov-Wasserstein Graph Mixup for Graph-level Classifications Fused Gromov-Wasserstein Graph Mixup for Graph-level Classifications NeurIPS 2023
- ICML 2023: Graph Mixup with Soft Alignments Graph Mixup with Soft Alignments ICML 2023
- KDD 2023: Improving Expressivity of GNNs with Subgraph-specific Factor Embedded Normalization Improving Expressivity of GNNs with Subgraph-specific Factor Embedded Normalization KDD 2023
- AAAI 2023: Union Subgraph Neural Networks Union Subgraph Neural Networks AAAI 2023
- GPT4Graph: Can Large Language Models Understand Graph Structured Data? An Empirical Evaluation and Benchmarking GPT4Graph: Can Large Language Models Understand Graph Structured Data? An Empirical Evaluation and Benchmarking
- ICML 2023: From Relational Pooling to Subgraph GNNs: A Universal Framework for More Expressive Graph Neural Networks From Relational Pooling to Subgraph GNNs: A Universal Framework for More Expressive Graph Neural Networks ICML 2023
- NeurIPS 2023: Graph Mixture of Experts: Learning on Large-Scale Graphs with Explicit Diversity Modeling Graph Mixture of Experts: Learning on Large-Scale Graphs with Explicit Diversity Modeling NeurIPS 2023
- KDD 2023: An Efficient Subgraph GNN with Provable Substructure Counting Power An Efficient Subgraph GNN with Provable Substructure Counting Power KDD 2023
- ICML 2023: A Complete Expressiveness Hierarchy for Subgraph GNNs via Subgraph Weisfeiler-Lehman Tests A Complete Expressiveness Hierarchy for Subgraph GNNs via Subgraph Weisfeiler-Lehman Tests ICML 2023
- ICML 2023: On the Connection Between MPNN and Graph Transformer On the Connection Between MPNN and Graph Transformer ICML 2023
- LoG 2022: You Can Have Better Graph Neural Networks by Not Training Weights at All: Finding Untrained GNNs Tickets You Can Have Better Graph Neural Networks by Not Training Weights at All: Finding Untrained GNNs Tickets LoG 2022
- Comprehensive Analysis of Over-smoothing in Graph Neural Networks from Markov Chains Perspective Comprehensive Analysis of Over-smoothing in Graph Neural Networks from Markov Chains Perspective
- ICLR 2022: Boosting the Cycle Counting Power of Graph Neural Networks with I$^2$-GNNs Boosting the Cycle Counting Power of Graph Neural Networks with I$^2$-GNNs ICLR 2022
- NeurIPS 2022: Geodesic Graph Neural Network for Efficient Graph Representation Learning Geodesic Graph Neural Network for Efficient Graph Representation Learning NeurIPS 2022
- LoG 2022: Expander Graph Propagation Expander Graph Propagation LoG 2022
- Graph Contrastive Learning with Personalized Augmentation Graph Contrastive Learning with Personalized Augmentation
- Motif-based Graph Representation Learning with Application to Chemical Molecules Motif-based Graph Representation Learning with Application to Chemical Molecules
- ICLR 2022: Agent-based Graph Neural Networks Agent-based Graph Neural Networks ICLR 2022
- KDD 2022: Condensing Graphs via One-Step Gradient Matching Condensing Graphs via One-Step Gradient Matching KDD 2022
- KDD 2022: Graph Rationalization with Environment-based Augmentations Graph Rationalization with Environment-based Augmentations KDD 2022
- ICML 2022: Omni-Granular Ego-Semantic Propagation for Self-Supervised Graph Representation Learning Omni-Granular Ego-Semantic Propagation for Self-Supervised Graph Representation Learning ICML 2022
- KDD 2022: Compressing Deep Graph Neural Networks via Adversarial Knowledge Distillation Compressing Deep Graph Neural Networks via Adversarial Knowledge Distillation KDD 2022
- Embedding Graphs on Grassmann Manifold Embedding Graphs on Grassmann Manifold
- ICLR 2022: Automated Data Augmentations for Graph Classification Automated Data Augmentations for Graph Classification ICLR 2022
- Transformer for Graphs: An Overview from Architecture Perspective Transformer for Graphs: An Overview from Architecture Perspective
- ICML 2022: Molecular Representation Learning via Heterogeneous Motif Graph Neural Networks Molecular Representation Learning via Heterogeneous Motif Graph Neural Networks ICML 2022
- GRPE: Relative Positional Encoding for Graph Transformer GRPE: Relative Positional Encoding for Graph Transformer
- ICLR 2022: Discovering Invariant Rationales for Graph Neural Networks Discovering Invariant Rationales for Graph Neural Networks ICLR 2022
- TMLR 2022: How Expressive are Transformers in Spectral Domain for Graphs? How Expressive are Transformers in Spectral Domain for Graphs? TMLR 2022
- SoGCN: Second-Order Graph Convolutional Networks SoGCN: Second-Order Graph Convolutional Networks
- NeurIPS 2021: Reconstruction for Powerful Graph Representations Reconstruction for Powerful Graph Representations NeurIPS 2021
- ICLR 2021: Do We Need Anisotropic Graph Neural Networks? Do We Need Anisotropic Graph Neural Networks? ICLR 2021
- Parameterized Hypercomplex Graph Neural Networks for Graph Classification Parameterized Hypercomplex Graph Neural Networks for Graph Classification
- Multi-Level Attention Pooling for Graph Neural Networks: Unifying Graph Representations with Multiple Localities Multi-Level Attention Pooling for Graph Neural Networks: Unifying Graph Representations with Multiple Localities
- ICML 2021: How Framelets Enhance Graph Neural Networks How Framelets Enhance Graph Neural Networks ICML 2021
- Graph convolutions that can finally model local structure Graph convolutions that can finally model local structure
- ICML 2020: GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training ICML 2020
- AAAI 2020: Rethinking Graph Regularization for Graph Neural Networks Rethinking Graph Regularization for Graph Neural Networks AAAI 2020
- ICLR 2020: Wasserstein Embedding for Graph Learning Wasserstein Embedding for Graph Learning ICLR 2020
Binary graph classification Predict whether a compound inhibits HIV replication. Evaluated by ROC-AUC.
Dataset details
Task
Predict whether a compound inhibits HIV replication. This is one binary graph-classification task scored by ROC-AUC because active compounds are rare.
Data
The 41,127 compounds are OGB’s RDKit-processed version of the MoleculeNet HIV dataset. Each graph is one molecule, with atoms as nodes and covalent bonds as edges; the activity labels ultimately come from the Drug Therapeutics Program AIDS antiviral screen.
Features
OGB converts SMILES into nine categorical atom fields, including atomic number, chirality, formal charge, aromaticity, and ring membership, and three categorical bond fields describing bond type, stereochemistry, and conjugation.
Splits and evaluation
The official split separates compounds by Bemis-Murcko-style two-dimensional molecular scaffolds rather than assigning molecules independently at random. ROC-AUC is computed on the held-out scaffold groups.
Quirks and pitfalls
The assay is highly imbalanced and a random split is easier because close structural analogues can occur on both sides. The label reports activity in a particular screening pipeline, not clinical efficacy, binding mechanism, or general antiviral action. RDKit parsing and OGB feature versions should be held fixed for comparison.
Sources
OGB scaffold split; use the official default task and features.
SOTA Progression
new SOTA
matches SOTA at that time
other proposed result
other baseline result
Leaderboard
| # | Model | ROC-AUC ↑ | AUC | AUC (MolHiv column) | AUC-ROC | AUROC | Accuracy | ROC-AUC (AUROC, percent form) | ROC-AUC (percentage) | Size | Sparsity | Suffic. | Paper |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Multi-RF Fusion with Multi-GNN Blending | 0.8476
±
0.0002 | — | — | — | — | — | — | — | — | — | — | |
| 2 | HyperFusion | 0.8475
±
0.0003 | — | — | — | — | — | — | — | — | — | — | |
| 3 | PAS+FPs | 0.8420
±
0.0015 | — | — | — | — | — | — | — | — | — | — | |
| 4 | SigGate-GT | 0.8247
±
0.0063 | — | — | — | — | — | — | — | — | — | — | |
| 5 | GRPE-Standard | 0.8139
±
0.0049 | — | — | — | — | — | — | — | — | — | — | |
| 6 | GSN + Granola | 0.8112
±
0.0079 | — | — | — | — | — | — | — | — | — | — | |
| 7 | GRM | 0.8102 | — | — | — | — | — | — | — | — | — | — | |
| 8 | CIN | 0.8094
±
0.0057 | — | — | — | — | — | — | — | — | — | — | |
| 9 | GatedGCN | 0.8090
±
0.0160 | — | — | — | — | — | — | — | — | — | — | |
| 10 | Exphormer | 0.8075
±
0.0094 | — | — | — | — | — | — | — | — | — | — | |
| 11 | EGT-Large | 0.8060
±
0.0065 | — | — | — | — | — | — | — | — | — | — | |
| 12 | GraphGPS FT | 0.8060
±
0.0070 | — | — | — | — | — | — | — | — | — | — | |
| 13 | GALA | 0.8053 | — | — | — | — | — | — | — | — | — | — | |
| 14 | Graphormer-FLAG | 0.8051
±
0.0053 | — | — | — | — | — | — | — | — | — | — | |
| 15 | Graphormer | 0.8051
±
0.0053 | — | — | — | — | — | — | — | — | — | — | |
| 16 | Graphormer FT | 0.8050
±
0.0050 | — | — | — | — | — | — | — | — | — | — | |
| 17 | Graphormer DeepGPT | 0.8040
±
0.0210 | — | — | — | — | — | — | — | — | — | — | |
| 18 | GSN | 0.8039
±
0.0090 | — | — | — | — | — | — | — | — | — | — | |
| 19 | DGSN | 0.8039
±
0.0090 | — | — | — | — | — | — | — | — | — | — | |
| 20 | Directional GSN | 0.8039
±
0.0090 | — | — | — | — | — | — | — | — | — | — | |
| 21 | GIN + DiGRAF | 0.8028
±
0.0144 | — | — | — | — | — | — | — | — | — | — | |
| 22 | GPS + RandAlign | 0.8021
±
0.0305 | — | — | — | — | — | — | — | — | — | — | |
| 23 | A-NLSF | 0.8019
±
0.0152 | — | — | — | — | — | — | — | — | — | — | |
| 24 | GSAT | 0.8018
±
0.0182 | — | — | — | — | — | — | — | — | — | — | |
| 25 | PF-GNN | 0.8015
±
0.0068 | — | — | — | — | — | — | — | — | — | — | |
| 26 | GraphGPS DeepGPT | 0.8010
±
0.0150 | — | — | — | — | — | — | — | — | — | — | |
| 27 | CIN-Small | 0.8005
±
0.0104 | — | — | — | — | — | — | — | — | — | — | |
| 28 | SUN (EGO) | 0.8003
±
0.0055 | — | — | — | — | — | — | — | — | — | — | |
| 29 | SUN(EGO) | 0.8003
±
0.0055 | — | — | — | — | — | — | — | — | — | — | |
| 30 | SUN | 0.8003
±
0.0055 | — | — | — | — | — | — | — | — | — | — | |
| 31 | LiGhT DeepGPT | 0.7990
±
0.0100 | — | — | — | — | — | — | — | — | — | — | |
| 32 | EGTAS | 0.7981
±
0.0117 | — | — | — | — | — | — | — | — | — | — | |
| 33 | GECO | 0.7980
±
0.0200 | — | — | — | — | — | — | — | — | — | — | |
| 34 | DLG | 0.7980
±
0.0178 | — | — | — | — | — | — | — | — | — | — | |
| 35 | PiPE | 0.7980
±
0.0040 | — | — | — | — | — | — | — | — | — | — | |
| 36 | CIGA | 0.7970
±
0.0192 | — | — | — | — | — | — | — | — | — | — | |
| 37 | DGN | 0.7970
±
0.0097 | — | — | — | — | — | — | — | — | — | — | |
| 38 | GNNAK+ | 0.7961
±
0.0119 | — | — | — | — | — | — | — | — | — | — | |
| 39 | AT Mask-1 | 0.7960 | — | — | — | — | — | — | — | — | — | — | |
| 40 | GNN-SSWL+ | 0.7958
±
0.0035 | — | — | — | — | — | — | — | — | — | — | |
| 41 | SSWL+ | 0.7958
±
0.0035 | — | — | — | — | — | — | — | — | — | — | |
| 42 | GraphAKD | 0.7946
±
0.0097 | — | — | — | — | — | — | — | — | — | — | |
| 43 | PIN | 0.7944
±
0.0140 | — | — | — | — | — | — | — | — | — | — | |
| 44 | DeeperGCN-FLAG | 0.7942
±
0.0120 | — | — | — | — | — | — | — | — | — | — | |
| 45 | DeeperGCN+FLAG | 0.7942
±
0.0120 | — | — | — | — | — | — | — | — | — | — | |
| 46 | GIN + EGP | 0.7934
±
0.0035 | — | — | — | — | — | — | — | — | — | — | |
| 47 | PHC-4 | 0.7934
±
0.0116 | — | — | — | — | — | — | — | — | — | — | |
| 48 | PHC-GNN | 0.7934
±
0.0116 | — | — | — | — | — | — | — | — | — | — | |
| 49 | GREA | 0.7932
±
0.0092 AUC
0.7932
±
0.0092 | 0.7932
±
0.0092 | — | — | — | — | — | — | — | — | — | |
| 50 | 4-IDMPNN | 0.7931
±
0.0063 | — | — | — | — | — | — | — | — | — | — |
SOTA performance
set a new SOTA at the time
matched the SOTA at the time