A hop-ordering contrastive loss for graphs outperforms existing self-supervised baselines on node classification, but the accompanying random-walk proof of universal label-consistency decay is misaligned with the empirical measure.
Momentum contrast for unsu- pervised visual representation learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Rethinking Graph Contrastive Learning through Relative Similarity Preservation
A hop-ordering contrastive loss for graphs outperforms existing self-supervised baselines on node classification, but the accompanying random-walk proof of universal label-consistency decay is misaligned with the empirical measure.