Latent cluster correction, a Louvain-based clustering loss, improves CIFAR-100 fine-tuning accuracy by up to 1 point on ResNet-18, but not on AlexNet.
Separability and Geometry of Object Manifolds in Deep Neural Networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Improving Fine-Tuning with Latent Cluster Correction
Latent cluster correction, a Louvain-based clustering loss, improves CIFAR-100 fine-tuning accuracy by up to 1 point on ResNet-18, but not on AlexNet.