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

In this study, we explore Transformer-based diffusion models for image andvideo generation. Despite the dominance of Transformer architectures in variousfields due to their flexibility and scalability, the visual generative domainprimarily utilizes CNN-based U-Net architectures, particularly indiffusion-based models. We introduce GenTron, a family of Generative modelsemploying Transformer-based diffusion, to address this gap. Our initial stepwas to adapt Diffusion Transformers (DiTs) from class to text conditioning, aprocess involving thorough empirical exploration of the conditioning mechanism.We then scale GenTron from approximately 900M to over 3B parameters, observingsignificant improvements in visual quality. Furthermore, we extend GenTron totext-to-video generation, incorporating novel motion-free guidance to enhancevideo quality. In human evaluations against SDXL, GenTron achieves a 51.1% winrate in visual quality (with a 19.8% draw rate), and a 42.3% win rate in textalignment (with a 42.9% draw rate). GenTron also excels in the T2I-CompBench,underscoring its strengths in compositional generation. We believe this workwill provide meaningful insights and serve as a valuable reference for futureresearch.

GenTron: Delving Deep into Diffusion Transformers for Image and Video Generation