Research map / Graph learning / GNN architectures
gHAWK: Structural Encoding for Scalable Training of Graph Neural Networks on Knowledge Graphs
Humera Sabir, Fatima Farooq, Ashraf Aboulnaga
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes gHAWK, a structural-encoding approach for scalable GNN training on knowledge graphs that avoids reliance on iterative message passing to learn graph structure.
Problem
Existing message-passing GNNs struggle to scale to large knowledge graphs because learning graph structure through iterative message passing is inefficient, especially under mini-batch training where nodes see only partial neighborhoods.
Approach
Encodes graph structure explicitly through structural encodings rather than relying on iterative message passing to learn it, targeting efficient mini-batch training on large KGs.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Bloom filter and TransE structural features on top of default text features | six GNN backbones · OGB-MAG and MAG240M | test accuracy: OGB-MAG +9.5 (ClusterGCN) to +17.2 (GAT); MAG240M +5.0 (GraphSAGE) to +16.8 (HGT) | Improved |
| add Bloom filter alone (no TransE) to default features (ablation) | ClusterGCN · MAG240M | test accuracy: 61.13 -> 57.37 | Hurt |
| add Bloom filter alone (no TransE) to default features (ablation) | R-GCN, GraphSAGE, GraphSAINT, GAT, HGT · OGB-MAG | test accuracy: GraphSAINT 46.81 -> 50.26; R-GCN 37.67 -> 44.51 | Improved |
| add TransE on top of Bloom filter (ablation) | GraphSAINT · OGB-MAG | test accuracy: 50.26 -> 57.37 | Improved |
| replace default text features with random vectors plus Bloom and TransE (ablation) | six GNN backbones · OGB-MAG | test accuracy: GraphSAINT 46.81 (Word2Vec only) -> 54.55 (structure only) | Improved |
| replace RoBERTa features with structure-only gHAWK features (ablation) | six GNN backbones · MAG240M | test accuracy: R-GCN 68.00 (RoBERTa) vs 59.23 (structure only) | Mixed |
| add gHAWK features to a message-passing-free MLP | MLP · OGB-MAG | test accuracy: 31.95 -> 35.61 | Improved |
| add gHAWK features to SGC, SIGN and SGFormer | SGC, SIGN, SGFormer · OGB-MAG | test accuracy: SGFormer 51.35 -> 56.46; SGC 38.42 -> 40.84; SIGN 37.54 -> 40.46 | Improved |
| add gHAWK features to GraphSAGE | GraphSAGE · DBLP, ACM, IMDb (HGB) | Micro/Macro-F1: DBLP Micro-F1 82.50 -> 97.52; ACM Macro-F1 91.20 -> 93.16; IMDb Micro-F1 65.28 -> 69.14 | Improved |
| add gHAWK structural features to RotatE decoder (no message passing) | RotatE · OGB-WikiKG2 | test MRR: 0.4341 -> 0.6941 | Improved |
| add gHAWK features to message-passing backbones | R-GCN, GraphSAGE, CompGCN · OGB-WikiKG2 | test MRR: R-GCN 0.3851 -> 0.5957; GraphSAGE 0.5689 -> 0.6129; CompGCN 0.6256 -> 0.6562 | Improved |
| add gHAWK features to RotatE, by head-entity training frequency | RotatE · OGB-WikiKG2 head queries (single run) | head MRR: 1-5 triples 0.0109 -> 0.3497; 1000+ triples 0.8989 -> 0.7380 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RotatE — RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space (ICLR 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Graph Neural Networks Powered by Encoder Embedding for Improved Node Learning (TPAMI 2026 · similarity 0.91)
- VQGraph: Rethinking Graph Representation Space for Bridging GNNs and MLPs (ICLR 2024 · similarity 0.90)
- Large Language Model Meets Graph Neural Network in Knowledge Distillation (AAAI 2025 · similarity 0.89)
- Contextual Structure Knowledge Transfer for Graph Neural Networks (AAAI 2025 · similarity 0.89)
- Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence (ICML 2025 · similarity 0.89)
- UniGTE: Unified Graph–Text Encoding for Zero-Shot Generalization across Graph Tasks and Domains (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
