English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Class-incremental learning aims to learn new classes in an incrementalfashion without forgetting the previously learned ones. Several research workshave shown how additional data can be used by incremental models to helpmitigate catastrophic forgetting. In this work, following the recentbreakthrough in text-to-image generative models and their wide distribution, wepropose the use of a pretrained Stable Diffusion model as a source ofadditional data for class-incremental learning. Compared to competitive methodsthat rely on external, often unlabeled, datasets of real images, our approachcan generate synthetic samples belonging to the same classes as the previouslyencountered images. This allows us to use those additional data samples notonly in the distillation loss but also for replay in the classification loss.Experiments on the competitive benchmarks CIFAR100, ImageNet-Subset, andImageNet demonstrate how this new approach can be used to further improve theperformance of state-of-the-art methods for class-incremental learning on largescale datasets.

Class-Incremental Learning using Diffusion Model for Distillation and Replay