Research map / Graph learning / Knowledge and temporal graphs
ParamE: Regarding Neural Network Parameters as Relation Embeddings for Knowledge Graph Completion
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 |
|---|---|---|---|
| feed relation embedding as network parameters instead of as input (ParamE-CNN vs ConvE) (ablation) | ConvE-like CNN · FB15k-237 | MRR: 0.325 -> 0.393 | Helped |
| feed relation embedding as network parameters instead of as input (ParamE-CNN vs ConvE) (ablation) | ConvE-like CNN · WN18RR | MRR: 0.43 -> 0.461 | Helped |
| use gate structure instead of MLP as the relation-parameterized network (ablation) | ParamE · FB15k-237 | MRR: 0.314 -> 0.399 | Helped |
| use gate structure instead of CNN as the relation-parameterized network (ablation) | ParamE · WN18RR | MRR: 0.461 -> 0.489 | Helped |
| use gate structure instead of CNN as the relation-parameterized network (ablation) | ParamE · FB15k-237 | MRR: 0.393 -> 0.399 | Helped |
| ParamE-Gate vs SACN | KG embedding · FB15k-237 | Hits@1: 0.27 -> 0.310 | Helped |
| ParamE-Gate vs best baseline | KG embedding · WN18RR | Hits@10: 0.571 (RotatE) vs 0.538 | Hurt |
| larger relation training set size | ParamE-Gate · per-relation WN18RR and FB15k-237 | MRR: hypernym 34796 triplets MRR 0.196; verb group 1138 triplets MRR 0.975 | No clear effect |
Closest papers by meaning
- Convolutional 2D Knowledge Graph Embeddings (AAAI 2018 · similarity 0.90)
- Contextual Parameter Generation for Knowledge Graph Link Prediction (AAAI 2020 · similarity 0.89)
- KGDM: A Diffusion Model to Capture Multiple Relation Semantics for Knowledge Graph Embedding (AAAI 2024 · similarity 0.87)
- Replacing Paths with Connection-Biased Attention for Knowledge Graph Completion (AAAI 2025 · similarity 0.87)
- Knowledge Graph Completion with Relation-Aware Anchor Enhancement (AAAI 2025 · similarity 0.87)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.87)
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.