Research map / Graph learning / Knowledge and temporal graphs
Representation Learning of Knowledge Graphs with Entity Descriptions
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 |
|---|---|---|---|
| combine CNN description representations with TransE | TransE · FB15K entity prediction | Hits@10 (Filter): 66.1 -> 67.4 | Helped |
| combine CNN description representations with TransE | TransE · FB15K entity prediction | Mean Rank (Filter): 119 -> 91 | Helped |
| use description-only CNN representations instead of TransE | TransE · FB15K entity prediction | Hits@10 (Filter): 66.1 -> 57.6 | Hurt |
| use CNN encoder instead of CBOW (ablation) | DKRL · FB20K zero-shot entity prediction | Hits@10: 25.3 -> 29.5 | Helped |
| use CNN encoder instead of CBOW (ablation) | DKRL · FB20K zero-shot relation prediction | Hits@1: 50.3 -> 58.2 | Helped |
| use CNN encoder instead of CBOW (ablation) | DKRL · FB15K entity classification | MAP: 89.3 -> 90.1 | Helped |
| use description-based representations for all entities instead of structure-based for in-KG entities (ablation) | DKRL(CNN) · FB20K zero-shot entity prediction | Hits@10: 24.8 -> 29.5 | Helped |
| DKRL(CNN) vs BOW features | entity classification · FB20K zero-shot | MAP: 57.5 -> 61.9 | Helped |
Closest papers by meaning
- Entity-Agnostic Representation Learning for Parameter-Efficient Knowledge Graph Embedding (AAAI 2023 · similarity 0.90)
- Interpreting Knowledge Graph Relation Representation from Word Embeddings (ICLR 2021 · similarity 0.90)
- Toward a Unified Statistical Theory of Unsupervised Pretraining and Supervised Neural Knowledge Graph Learning (NeurIPS 2026 · similarity 0.88)
- Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation (Abstract Reprint) (AAAI 2024 · similarity 0.87)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.87)
- Replacing Paths with Connection-Biased Attention for 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.