Research map / Graph learning / Knowledge and temporal graphs
ConE: Cone Embeddings for Multi-Hop Reasoning over Knowledge Graphs
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 |
|---|---|---|---|
| replace Beta-distribution embeddings with cone embeddings | BetaE · FB15k EPFO | MRR: 41.6 -> 49.8 | Helped |
| replace Beta-distribution embeddings with cone embeddings | BetaE · FB15k negation queries | MRR: 11.8 -> 14.8 | Helped |
| replace Beta-distribution embeddings with cone embeddings | BetaE · FB15k-237 negation queries | MRR: 5.4 -> 5.9 | Helped |
| replace box embeddings (Q2B) with cone embeddings | Query2Box · FB15k EPFO | MRR: 40.1 -> 49.8 | Helped |
| implement union with De Morgan's law instead of DNF (ablation) | ConE · FB15k 2u | MRR: 55.7 -> 37.7 | Hurt |
| implement union with De Morgan's law instead of DNF (ablation) | ConE · FB15k up | MRR: 31.4 -> 29.8 | Hurt |
| implement union with De Morgan's law instead of DNF (ablation) | ConE · FB237 up | MRR: 10.8 -> 9.9 | Hurt |
| use cone apertures to model answer-set cardinality vs BetaE | ConE · FB15k 3p | Spearman correlation: 0.47 -> 0.70 | Helped |
| cone embeddings vs EmQL | ConE · FB15k generalization (EmQL setting) | MRR: 43.9 -> 52.9 | Helped |
Closest papers by meaning
- Query2box: Reasoning over Knowledge Graphs in Vector Space Using Box Embeddings (ICLR 2020 · similarity 0.91)
- Query Embedding on Hyper-Relational Knowledge Graphs (ICLR 2022 · similarity 0.91)
- Beta Embeddings for Multi-Hop Logical Reasoning in Knowledge Graphs (NeurIPS 2020 · similarity 0.90)
- NQE: N-ary Query Embedding for Complex Query Answering over Hyper-Relational Knowledge Graphs (AAAI 2023 · similarity 0.90)
- Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones (NeurIPS 2021 · similarity 0.90)
- Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors (ICLR 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.