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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Generative models have been very popular in the recent years for their imagegeneration capabilities. GAN-based models are highly regarded for theirdisentangled latent space, which is a key feature contributing to their successin controlled image editing. On the other hand, diffusion models have emergedas powerful tools for generating high-quality images. However, the latent spaceof diffusion models is not as thoroughly explored or understood. Existingmethods that aim to explore the latent space of diffusion models usually relieson text prompts to pinpoint specific semantics. However, this approach may berestrictive in areas such as art, fashion, or specialized fields like medicine,where suitable text prompts might not be available or easy to conceive thuslimiting the scope of existing work. In this paper, we propose an unsupervisedmethod to discover latent semantics in text-to-image diffusion models withoutrelying on text prompts. Our method takes a small set of unlabeled images fromspecific domains, such as faces or cats, and a pre-trained diffusion model, anddiscovers diverse semantics in unsupervised fashion using a contrastivelearning objective. Moreover, the learned directions can be appliedsimultaneously, either within the same domain (such as various types of facialedits) or across different domains (such as applying cat and face edits withinthe same image) without interfering with each other. Our extensive experimentsshow that our method achieves highly disentangled edits, outperforming existingapproaches in both diffusion-based and GAN-based latent space editing methods.

NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models