DIET-CP continues pretraining a vision model by classifying each image by its index on roughly 1000 unlabeled images, improving medical and astronomical classification while degrading fine-grained natural-image classification.
Efficient domain continual pretraining by mitigating the stability gap
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.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
DIET-CP: Lightweight and Data Efficient Self Supervised Continued Pretraining
DIET-CP continues pretraining a vision model by classifying each image by its index on roughly 1000 unlabeled images, improving medical and astronomical classification while degrading fine-grained natural-image classification.