The attention-based transformer architecture is powerful because it can process information differently based on various inputs, similar to how humans perceive concepts through different lenses. This flexibility is a significant advantage over simpler models like image classifiers. However, the quadratic nature of transformers makes them slow, as every additional token requires a relationship computation with all previous tokens. Despite attempts to optimize and approximate this process, the full dense non-approximated attention remains quadratic, with current approximations generally being less effective than the full attention.

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