Rethinking Class-incremental Learning in the Era of Large Pre-trained Models via Test-Time Adaptation

arXiv:2310.11482v2 Announce Type: replace
Abstract: Class-incremental learning (CIL) is a challenging task that involves sequentially learning to categorize classes from new tasks without forgetting previously learned information. The advent of large pre-trained models (PTMs) has fast-tracked the progress in CIL due to the highly transferable PTM representations, where tuning a small set of parameters leads to state-of-the-art performance when compared with the traditional CIL methods that are trained from scratch. However, repeated fine-tuning on each task destroys the rich representations …

arxiv class cs.ai cs.cv incremental pre-trained models test type via