MERA, a self-supervised Vision Transformer pipeline with sparse-seeded active learning, reports 86.2% malignancy accuracy on LIDC with roughly 1% of training labels, close to its 87.6% fully supervised accuracy, while producing global, case-based, visual, and concept explanations.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MERA: Multimodal and Multiscale Self-Explanatory Model with Considerably Reduced Annotation for Lung Nodule Diagnosis
MERA, a self-supervised Vision Transformer pipeline with sparse-seeded active learning, reports 86.2% malignancy accuracy on LIDC with roughly 1% of training labels, close to its 87.6% fully supervised accuracy, while producing global, case-based, visual, and concept explanations.