Research map / Graph learning / Knowledge and temporal graphs
Entity-Agnostic Representation Learning for Parameter-Efficient Knowledge Graph Embedding
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 EARL entity-agnostic encoding instead of conventional entity embeddings | RotatE · FB15k-237 | MRR: 0.296 -> 0.310 | Helped |
| Use EARL entity-agnostic encoding instead of conventional entity embeddings | RotatE · WN18RR | MRR: 0.411 -> 0.440 | Helped |
| Use EARL entity-agnostic encoding instead of conventional entity embeddings | RotatE · CoDEx-L | MRR: 0.196 -> 0.238 | Helped |
| Use EARL entity-agnostic encoding instead of conventional entity embeddings | RotatE · YAGO3-10 | MRR: 0.121 -> 0.302 | Helped |
| Add reserved entities and their embeddings (ablation) | EARL · FB15k-237, WN18RR, CoDEx-L, YAGO3-10 | MRR: 0.306 -> 0.310; 0.347 -> 0.440; 0.203 -> 0.238; 0.119 -> 0.302 | Mixed |
| Add ConRel relation information (ablation) | EARL · FB15k-237, WN18RR, CoDEx-L, YAGO3-10 | MRR: 0.309 -> 0.310; 0.432 -> 0.440; 0.237 -> 0.238; 0.322 -> 0.302 | Mixed |
| Add kNResEnt information (ablation) | EARL · FB15k-237, WN18RR, CoDEx-L, YAGO3-10 | MRR: 0.301 -> 0.310; 0.409 -> 0.440; 0.232 -> 0.238; 0.249 -> 0.302 | Helped |
| Use ConRel and kNResEnt instead of random entity representations (ablation) | EARL · FB15k-237, WN18RR, CoDEx-L, YAGO3-10 | MRR: 0.302 -> 0.310; 0.350 -> 0.440; 0.234 -> 0.238; 0.286 -> 0.302 | Helped |
| Add MulHop neighbor encoding (ablation) | EARL · FB15k-237, WN18RR, CoDEx-L, YAGO3-10 | MRR: 0.250 -> 0.310; 0.048 -> 0.440; 0.095 -> 0.238; 0.033 -> 0.302 | Helped |
| Vary k for retrieving nearest reserved entities (ablation) | EARL · FB15k-237 and WN18RR | MRR: about 0.01 | No clear effect |
| Increase k beyond a middle value (ablation) | EARL · FB15k-237 and WN18RR | MRR | Mixed |
Closest papers by meaning
- Representation Learning of Knowledge Graphs with Entity Descriptions (AAAI 2016 · similarity 0.90)
- NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs (ICLR 2022 · similarity 0.89)
- ParamE: Regarding Neural Network Parameters as Relation Embeddings for Knowledge Graph Completion (AAAI 2020 · similarity 0.89)
- TransGate: Knowledge Graph Embedding with Shared Gate Structure (AAAI 2019 · similarity 0.88)
- Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation (Abstract Reprint) (AAAI 2024 · similarity 0.87)
- KGDM: A Diffusion Model to Capture Multiple Relation Semantics for Knowledge Graph Embedding (AAAI 2024 · 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.