State-of-the-art diffusion models have self-attention at their core for performance. Patchifying images before applying self-attention is deemed detrimental, as it leads to worse image quality. By using state-space models instead of self-attention, there is no need to compress image representations, enabling the models to handle longer images efficiently. This approach yields better scaling for longer images in image generation tasks, showcasing state-of-the-art performance on ImageNet.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode