LLMs represent semantic relations geometrically via embedding distance and direction; a linear Polar Probe decodes these structures from middle-layer activations and generalizes to new entities.
A review of relational machine learning for knowledge graphs
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
verdicts
UNVERDICTED 4representative citing papers
The paper formalizes query answering with soft constraints on knowledge graphs and introduces two lightweight methods (parameter tuning or small neural network) to incorporate them while preserving original rankings.
Empirical comparison shows gradient-based explanations for GNN node similarities are actionable, consistent, and retain effects when sparsified, unlike mutual information explanations.
BioBLP is a modular embedding framework for multimodal biomedical KGs supporting heterogeneous attributes and missing data, with a pretraining strategy that improves results on drug-protein interaction prediction especially for low-degree entities.
citing papers explorer
-
Polar probe linearly decodes semantic structures from LLMs
LLMs represent semantic relations geometrically via embedding distance and direction; a linear Polar Probe decodes these structures from middle-layer activations and generalizes to new entities.
-
Interactive Query Answering on Knowledge Graphs with Soft Entity Constraints
The paper formalizes query answering with soft constraints on knowledge graphs and introduces two lightweight methods (parameter tuning or small neural network) to incorporate them while preserving original rankings.
-
Explaining Graph Neural Networks for Node Similarity on Graphs
Empirical comparison shows gradient-based explanations for GNN node similarities are actionable, consistent, and retain effects when sparsified, unlike mutual information explanations.
-
BioBLP: A Modular Framework for Learning on Multimodal Biomedical Knowledge Graphs
BioBLP is a modular embedding framework for multimodal biomedical KGs supporting heterogeneous attributes and missing data, with a pretraining strategy that improves results on drug-protein interaction prediction especially for low-degree entities.