Introduces integration, metastability, and dynamical stability index measures from layer activations and reports patterns distinguishing CIFAR-10 from CIFAR-100 difficulty plus early convergence signals across ResNet variants, DenseNet, MobileNetV2, VGG-16, and a Vision Transformer.
Adam: A method for stochastic optimization,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
LPLCv2 is a larger, more annotated dataset for fine-grained license plate legibility classification with a baseline model reaching 89.5% F1-score via a new training method and camera-contamination protocol.
citing papers explorer
-
Training Deep Visual Networks Beyond Loss and Accuracy Through a Dynamical Systems Approach
Introduces integration, metastability, and dynamical stability index measures from layer activations and reports patterns distinguishing CIFAR-10 from CIFAR-100 difficulty plus early convergence signals across ResNet variants, DenseNet, MobileNetV2, VGG-16, and a Vision Transformer.
-
LPLCv2: An Expanded Dataset for Fine-Grained License Plate Legibility Classification
LPLCv2 is a larger, more annotated dataset for fine-grained license plate legibility classification with a baseline model reaching 89.5% F1-score via a new training method and camera-contamination protocol.