DWCL anchors all contrastive pairs to the single highest-silhouette view and uses dual quality and discrepancy weights, reporting gains on eight multi-view clustering benchmarks.
Support-vector networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DWCL: Dual-Weighted Contrastive Learning for Multi-View Clustering
DWCL anchors all contrastive pairs to the single highest-silhouette view and uses dual quality and discrepancy weights, reporting gains on eight multi-view clustering benchmarks.