CLID-MU replaces the clean meta-dataset in meta-learning with an unsupervised cross-layer divergence metric, improving noisy-label and semi-supervised results on several benchmarks.
Title resolution pending
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
-
CLID-MU: Cross-Layer Information Divergence Based Meta Update Strategy for Learning with Noisy Labels
CLID-MU replaces the clean meta-dataset in meta-learning with an unsupervised cross-layer divergence metric, improving noisy-label and semi-supervised results on several benchmarks.