Research map / Graph learning / Knowledge and temporal graphs
Inductive Logical Query Answering in Knowledge Graphs
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use relational-structure reasoning with end-to-end query training instead of transductive entity embeddings | GNN-QE · FB15k-237 inductive test, 175% | Hits@10 avg_p (%): 1.3 -> 50.7 | Helped |
| represent entities inductively from relation-type tokens and answer queries inference-only | NodePiece-QE · FB15k-237 inductive test, 175% | Hits@10 avg_p (%): 10.1 -> 11.2 | Helped |
| add a message-passing GNN to NodePiece-QE (ablation) | NodePiece-QE · FB15k-237 inductive test, 175% | Hits@10 avg_p (%): 11.2 -> 28.6 | Helped |
| add a message-passing GNN to NodePiece-QE (ablation) | NodePiece-QE · WikiKG-QE test, 2M nodes | Hits@100 avg_p: 9.2 -> 10.1 | Helped |
| use relational-structure query reasoning rather than inference-only node representations | GNN-QE vs NodePiece-QE · FB15k-237 inductive test, 175% | Hits@10 avg_p (%): 11.2 -> 50.7 | Helped |
| increase inference-graph size to 550% of training graph size | GNN-based models · FB15k-237 inductive test | Hits@10 avg_p: 20 absolute Hits@10 points | Hurt |
| increase inference-graph size | NodePiece-QE without GNN · FB15k-237 inductive test | Hits@10 | No clear effect |
| increase graph size and the inference-to-training entity ratio | NodePiece-QE without GNN · Training queries on extended inference graphs | Hits@10 | Helped |
| add message passing to NodePiece-QE (ablation) | NodePiece-QE · Training queries on extended inference graphs | Hits@10 | Hurt |
| train on complex query patterns | GNN-QE · FB15k-237 inductive test, 175% | ROC AUC avg_p: 0.692 -> 0.973 | Helped |
Closest papers by meaning
- Query Embedding on Hyper-Relational Knowledge Graphs (ICLR 2022 · similarity 0.91)
- GraphOracle: Efficient Fully-Inductive Knowledge Graph Reasoning via Relation-Dependency Graphs (AAAI 2026 · similarity 0.91)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.90)
- Context-aware Inductive Knowledge Graph Completion with Latent Type Constraints and Subgraph Reasoning (AAAI 2025 · similarity 0.90)
- Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors (ICLR 2024 · similarity 0.89)
- A Foundation Model for Zero-shot Logical Query Reasoning (NeurIPS 2024 · 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.