Research map / Graph learning / Knowledge and temporal graphs
Convolutional 2D Knowledge Graph Embeddings
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 ConvE instead of the strongest listed baseline, ANALOGY | WN18 link prediction · WN18 | MRR: .942 -> .943 | Hurt |
| Use ConvE instead of the strongest listed baseline, ANALOGY | FB15k link prediction · FB15k | MRR: .725 -> .657 | Hurt |
| Use ConvE instead of the strongest listed baseline, ComplEx | WN18RR link prediction · WN18RR | MRR: .44 -> {.43} | Hurt |
| Use ConvE instead of the strongest listed baseline, ComplEx | FB15k-237 link prediction · FB15k-237 | MRR: .247 -> {\textbf{.325} | Helped |
| Use ConvE instead of the strongest listed baseline, ComplEx | YAGO3-10 link prediction · YAGO3-10 | MRR: .36 -> .44 | Helped |
| Use ConvE instead of DistMult | Countries S2 · Countries S2 | AUC-PR: 0.72\text{$\pm$}0.12 -> \textbf{0.99\text{$\pm$}0.01} | Helped |
| Use ConvE instead of DistMult | Countries S3 · Countries S3 | AUC-PR: 0.52\text{$\pm$}0.07 -> 0.86 \text{$\pm$}0.05 | Helped |
| Add hidden dropout (ablation) | ConvE on FB15k-237 · FB15k-237 | Hits@10: -0.044 $\pm$ 0.003 | Helped |
| Add input dropout (ablation) | ConvE on FB15k-237 · FB15k-237 | Hits@10: -0.022 $\pm$ 0.000 | Helped |
| Add 1-N scoring (ablation) | ConvE on FB15k-237 · FB15k-237 | Hits@10: -0.019 | Helped |
| Add feature map dropout (ablation) | ConvE on FB15k-237 · FB15k-237 | Hits@10: -0.013 $\pm$ 0.001 | Helped |
| Add label smoothing (ablation) | ConvE on FB15k-237 · FB15k-237 | Hits@10: -0.008 $\pm$ 0.000 | Helped |
Closest papers by meaning
- InteractE: Improving Convolution-Based Knowledge Graph Embeddings by Increasing Feature Interactions (AAAI 2020 · similarity 0.92)
- End-to-End Structure-Aware Convolutional Networks for Knowledge Base Completion (AAAI 2019 · similarity 0.92)
- ParamE: Regarding Neural Network Parameters as Relation Embeddings for Knowledge Graph Completion (AAAI 2020 · similarity 0.90)
- Distilling Knowledge From Graph Convolutional Networks (CVPR 2020 · similarity 0.89)
- Composition-based Multi-Relational Graph Convolutional Networks (ICLR 2020 · similarity 0.89)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 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.