A training-free message-passing framework with dynamic pruning that answers existential first-order logic queries on knowledge graphs using fuzzy symbolic states plus pretrained neural link scores.
Query2particles: Knowledge graph reasoning with particle embeddings
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Neural-Symbolic Message Passing with Dynamic Pruning
A training-free message-passing framework with dynamic pruning that answers existential first-order logic queries on knowledge graphs using fuzzy symbolic states plus pretrained neural link scores.