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

Although autoregressive models have achieved promising results on imagegeneration, their unidirectional generation process prevents the resultantimages from fully reflecting global contexts. To address the issue, we proposean effective image generation framework of Draft-and-Revise with ContextualRQ-transformer to consider global contexts during the generation process. As ageneralized VQ-VAE, RQ-VAE first represents a high-resolution image as asequence of discrete code stacks. After code stacks in the sequence arerandomly masked, Contextual RQ-Transformer is trained to infill the masked codestacks based on the unmasked contexts of the image. Then, ContextualRQ-Transformer uses our two-phase decoding, Draft-and-Revise, and generates animage, while exploiting the global contexts of the image during the generationprocess. Specifically. in the draft phase, our model first focuses ongenerating diverse images despite rather low quality. Then, in the revisephase, the model iteratively improves the quality of images, while preservingthe global contexts of generated images. In experiments, our method achievesstate-of-the-art results on conditional image generation. We also validate thatthe Draft-and-Revise decoding can achieve high performance by effectivelycontrolling the quality-diversity trade-off in image generation.

Draft-and-Revise: Effective Image Generation with Contextual RQ-Transformer