Semi-supervised learning (SSL) offers a robust framework for harnessing the potential of unannotated data. Traditionally, SSL mandates that all classes possess labeled instances.

Continual semantic segmentation aims to learn new classes while maintaining the information from the previous classes.

Despite a realistic problem, quantization on class-imbalanced data has not been studied.

Existing long-tailed recognition methods, aiming totrain class-balanced models from long-tailed data, generally assume the models would be evaluated on the uniform test class distribution.

Severe data imbalance naturally exists among web-scale vision-language datasets.