Research map / Graph learning / Knowledge and temporal graphs
Neural-Symbolic Models for Logical Queries on 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 fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · FB15k | avg_p MRR (%): 58.2 -> 72.8 | Helped |
| Use fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · FB15k | avg_n MRR (%): 14.8 -> 38.6 | Helped |
| Use fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · FB15k-237 | avg_p MRR (%): 24.0 -> 26.8 | Helped |
| Use fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · FB15k-237 | avg_n MRR (%): 7.8 -> 10.2 | Helped |
| Use fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · NELL995 | avg_p MRR (%): 28.8 -> 28.9 | Helped |
| Use fuzzy-set query execution with learned GNN projections and product fuzzy logic | FOL query answering · NELL995 | avg_n MRR (%): 7.8 -> 9.7 | Helped |
| Use traversal dropout to mask query-traversed edges during training (ablation) | GNN-QE · FB15k-237 validation EPFO queries | MRR | Helped |
| Set traversal dropout probability to 0, removing edge masking (ablation) | GNN-QE · FB15k-237 validation EPFO queries | MRR | Hurt |
| Set traversal dropout probability to 1, masking every traversed edge (ablation) | GNN-QE · FB15k-237 validation EPFO queries | MRR | Hurt |
| Replace NBFNet with RGCN for relation projection (ablation) | GNN-QE · FB15k-237 | avg_p MRR (%): 26.8 -> 20.9 | Hurt |
| Replace NBFNet with RGCN for relation projection (ablation) | GNN-QE · FB15k-237 | avg_n MRR (%): 10.2 -> 7.3 | Hurt |
| Replace NBFNet with CompGCN for relation projection (ablation) | GNN-QE · FB15k-237 | avg_p MRR (%): 26.8 -> 22.5 | Hurt |
Closest papers by meaning
- Neural Methods for Logical Reasoning over Knowledge Graphs (ICLR 2022 · similarity 0.92)
- Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors (ICLR 2024 · similarity 0.92)
- Fuzzy Logic Based Logical Query Answering on Knowledge Graphs (AAAI 2022 · similarity 0.91)
- Efficient Embeddings of Logical Variables for Query Answering over Incomplete Knowledge Graphs (AAAI 2023 · similarity 0.90)
- Neural-Symbolic Entangled Framework for Complex Query Answering (NeurIPS 2022 · similarity 0.90)
- Inductive Logical Query Answering in Knowledge Graphs (NeurIPS 2022 · similarity 0.89)
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.