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

In this work, we investigate the problem of Model-Agnostic Zero-ShotClassification (MA-ZSC), which refers to training non-specific classificationarchitectures (downstream models) to classify real images without using anyreal images during training. Recent research has demonstrated that generatingsynthetic training images using diffusion models provides a potential solutionto address MA-ZSC. However, the performance of this approach currently fallsshort of that achieved by large-scale vision-language models. One possibleexplanation is a potential significant domain gap between synthetic and realimages. Our work offers a fresh perspective on the problem by providing initialinsights that MA-ZSC performance can be improved by improving the diversity ofimages in the generated dataset. We propose a set of modifications to thetext-to-image generation process using a pre-trained diffusion model to enhancediversity, which we refer to as our $\textbf{bag of tricks}$. Our approachshows notable improvements in various classification architectures, withresults comparable to state-of-the-art models such as CLIP. To validate ourapproach, we conduct experiments on CIFAR10, CIFAR100, and EuroSAT, which isparticularly difficult for zero-shot classification due to its satellite imagedomain. We evaluate our approach with five classification architectures,including ResNet and ViT. Our findings provide initial insights into theproblem of MA-ZSC using diffusion models. All code will be available on GitHub.

Diversity is Definitely Needed: Improving Model-Agnostic Zero-shot Classification via Stable Diffusion