F²LP-AP is a fast training-free label propagation framework that adapts propagation parameters via local clustering coefficients and geometric medians to achieve competitive accuracy to GNNs on node classification while being computationally efficient.
In: Advances in Neural Information Processing Systems (NeurIPS) (2023)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
F\textsuperscript{2}LP-AP: Fast \& Flexible Label Propagation with Adaptive Propagation Kernel
F²LP-AP is a fast training-free label propagation framework that adapts propagation parameters via local clustering coefficients and geometric medians to achieve competitive accuracy to GNNs on node classification while being computationally efficient.