Research map / Label-efficient and robust learning / Recognition and long-tail
PolygonE: Modeling N-ary Relational Data as Gyro-Polygons in Hyperbolic Space
Shiyao Yan, Zequn Zhang, Xian Sun, Guangluan Xu, Shuchao Li, Qing Liu, Nayu Liu, Shensi Wang
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| model intact n-ary facts as hyperbolic gyro-polygons and score them using vertex-to-gyrocentroid distance | PolygonE · WikiPeople | MRR: .380 -> .431 | Improved |
| model intact n-ary facts as hyperbolic gyro-polygons and score them using vertex-to-gyrocentroid distance | PolygonE · JF17K | MRR: .539 -> .565 | Improved |
| model intact n-ary facts as hyperbolic gyro-polygons and score them using vertex-to-gyrocentroid distance | PolygonE · FB-AUTO | MRR: .830 -> .858 | Improved |
| use PolygonE for binary knowledge-base completion | PolygonE · WikiPeople | MRR: 0.451 -> 0.462 | Improved |
| use PolygonE for binary knowledge-base completion | PolygonE · JF17K | MRR: 0.348 -> 0.364 | Improved |
| use PolygonE for binary knowledge-base completion | PolygonE · FB-AUTO | MRR: 0.557 -> 0.565 | Improved |
| use PolygonE for binary knowledge-base completion | PolygonE · WN18 | MRR: 0.955 -> 0.953 | Hurt |
| use PolygonE for binary knowledge-base completion | PolygonE · FB15K | MRR: 0.850 -> 0.824 | Hurt |
| use vertex-gyrocentroid distance rather than polygon perimeter for optimization (ablation) | PolygonE · JF17K | MRR: .498 -> .565 | Improved |
| use vertex-gyrocentroid distance rather than polygon perimeter for optimization (ablation) | PolygonE · FB-AUTO | MRR: .824 -> .858 | Improved |
| use hyperbolic gyrovector operations instead of Euclidean equivalents (ablation) | PolygonE · JF17K | MRR: .548 -> .565 | Improved |
| use hyperbolic gyrovector operations instead of Euclidean equivalents (ablation) | PolygonE · FB-AUTO | MRR: .832 -> .858 | Improved |
Closest papers by meaning
- NestE: Modeling Nested Relational Structures for Knowledge Graph Reasoning (AAAI 2024 · similarity 0.87)
- Generative Representation Learning on Hyper-relational Knowledge Graphs via Masked Discrete Diffusion (ICML 2026 · similarity 0.86)
- ConE: Cone Embeddings for Multi-Hop Reasoning over Knowledge Graphs (NeurIPS 2021 · similarity 0.86)
- Text2NKG: Fine-Grained N-ary Relation Extraction for N-ary relational Knowledge Graph Construction (NeurIPS 2024 · similarity 0.85)
- HGE: Embedding Temporal Knowledge Graphs in a Product Space of Heterogeneous Geometric Subspaces (AAAI 2024 · similarity 0.84)
- Structure Is All You Need: Structural Representation Learning on Hyper-Relational Knowledge Graphs (ICML 2025 · similarity 0.84)
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.
