Research map / Large language models / Retrieval and knowledge
Neural Methods for Logical 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 |
|---|---|---|---|
| Use MLP rather than BetaE | KG reasoning on FB15k EPFO queries · FB15k | MRR (%): 41.6 -> 43.9 | Helped |
| Use MLP-Mixer rather than BetaE | KG reasoning on FB15k-237 EPFO queries · FB15k-237 | MRR (%): 20.9 -> 22.9 | Helped |
| Use MLP-Mixer rather than BetaE | KG reasoning on NELL995 EPFO queries · NELL995 | MRR (%): 24.6 -> 27.4 | Helped |
| Use MLP rather than BetaE for queries with negation | KG reasoning · FB15k negative queries | MRR (%): 11.8 -> 14.5 | Helped |
| Use MLP rather than BetaE for queries with negation | KG reasoning · FB15k-237 negative queries | MRR (%): 5.4 -> 6.9 | Helped |
| Use MLP rather than BetaE for queries with negation | KG reasoning · NELL995 negative queries | MRR (%): 5.9 -> 6.1 | Mixed |
| Use two separate networks and average their embeddings (ablation) | MLP · FB15k EPFO queries | MRR (%): 43.9 -> 45.8 | Helped |
| Add attention to the intersection operator (ablation) | MLP · FB15k EPFO queries | MRR (%): 43.9 -> 43.8 | No clear effect |
| Transform embeddings with heterogeneous hyper-graph embeddings (ablation) | MLP · FB15k EPFO queries | MRR (%): 43.9 -> 36.8 | Hurt |
| Use two separate networks and average their embeddings (ablation) | MLP · FB15k-237 EPFO queries | MRR (%): 22.6 -> 23.6 | Helped |
| Add attention to the intersection operator (ablation) | MLP · FB15k-237 EPFO queries | MRR (%): 22.6 -> 23.1 | Helped |
| Transform embeddings with heterogeneous hyper-graph embeddings (ablation) | MLP · FB15k-237 EPFO queries | MRR (%): 22.6 -> 20.6 | Hurt |
Closest papers by meaning
- Beta Embeddings for Multi-Hop Logical Reasoning in Knowledge Graphs (NeurIPS 2020 · similarity 0.94)
- Neural-Symbolic Models for Logical Queries on Knowledge Graphs (ICML 2022 · similarity 0.92)
- Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors (ICLR 2024 · similarity 0.92)
- Efficient Embeddings of Logical Variables for Query Answering over Incomplete Knowledge Graphs (AAAI 2023 · similarity 0.92)
- Query Embedding on Hyper-Relational Knowledge Graphs (ICLR 2022 · similarity 0.90)
- An Evaluation of Approaches to Train Embeddings for Logical Inference (Student Abstract) (AAAI 2025 · similarity 0.88)
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.