Research map / Graph learning / Knowledge and temporal graphs
TransGate: Knowledge Graph Embedding with Shared Gate Structure
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 |
|---|---|---|---|
| add shared gates over TransE (TransGate(fc)) | TransE · FB15K-237 | Hits@10 (%): 46.5 -> 58.1 | Helped |
| add shared gates over TransE (TransGate(fc)) | TransE · FB15K | Hits@10 (%): 47.1 -> 91.4 | Helped |
| add shared gates over TransE (TransGate(fc)) | TransE · WN18RR | Hits@10 (%): 50.1 -> 51.0 | No clear effect |
| replace fully connected gate with weight-vector gate (wv) (ablation) | TransGate · FB15K-237 | Hits@10 (%): 58.1 -> 54.2 | Hurt |
| replace fully connected gate with weight-vector gate (wv) (ablation) | TransGate · FB15K | Hits@10 (%): 91.4 -> 91.4 | No clear effect |
| replace fully connected gate with weight-vector gate (wv) (ablation) | TransGate · FB15K triplet classification | accuracy (%): 89.5 -> 89.3 | Hurt |
| replace fully connected gate with weight-vector gate (wv) (ablation) | TransGate · FB13 triplet classification | accuracy (%): 88.8 -> 88.1 | Hurt |
| replace fully connected gate with weight-vector gate (wv) (ablation) | TransGate · training time | minutes: 9.69min -> 5.11min | Helped |
| increase embedding size from 32 to 200 (ablation) | TransGate(fc) · FB15K-237 | Hits@10 (%): 41.5 -> 58.1 | Helped |
| increase embedding size from 32 to 200 (ablation) | TransGate(wv) · FB15K-237 | Hits@10 (%): 39.6 -> 54.2 | Helped |
| use TransGate(fc) instead of TransD | TransD · FB13 triplet classification | accuracy (%): 89.1 -> 88.8 | Hurt |
| TransGate(fc) at 0.74M parameters vs ConvE | ConvE · FB15K-237 | Hits@10 (%): 49.0 -> 51.7 | Helped |
Closest papers by meaning
- TorusE: Knowledge Graph Embedding on a Lie Group (AAAI 2018 · similarity 0.90)
- Locally Adaptive Translation for Knowledge Graph Embedding (AAAI 2016 · similarity 0.90)
- Relational Pattern Benchmarking on the Knowledge Graph Link Prediction Task (NeurIPS 2021 · similarity 0.89)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.88)
- KGDM: A Diffusion Model to Capture Multiple Relation Semantics for Knowledge Graph Embedding (AAAI 2024 · similarity 0.87)
- Toward a Unified Statistical Theory of Unsupervised Pretraining and Supervised Neural Knowledge Graph Learning (NeurIPS 2026 · 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.