Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
The Rise Of Large Database Models
 Originally published in Forbes Even as large language models have...
3 Predictions For Predictive AI In 2025
 Originally published in Forbes GenAI’s complementary sibling, predictive AI, makes...
The Quant’s Dilemma: Subjectivity In Predictive AI’s Value
 Originally published in Forbes This is the third of a...
To Deploy Predictive AI, You Must Navigate These Tradeoffs
 Originally published in Forbes This is the second of a...
SHARE THIS:

3 years ago
DALL·E 2, Explained: The Promise and Limitations of a Revolutionary AI

 
Originally published in Towards Data Science, June 16, 2022.

DALL·E 2 is the newest AI model by OpenAI. If you’ve seen some of its creations and think they’re amazing, keep reading to understand why you’re totally right — but also wrong.

OpenAI published a blog post and a paper entitled “Hierarchical Text-Conditional Image Generation with CLIP Latents” on DALL·E 2. The post is fine if you want to get a glimpse at the results and the paper is great for understanding the technical details, but neither explains DALL·E 2’s amazingness — and the not-so-amazing — in depth. That’s what this article is for.

DALL·E 2 is the new version of DALL·E, a generative language model that takes sentences and creates corresponding original images. At 3.5B parameters, DALL·E 2 is a large model but not nearly as large as GPT-3 and, interestingly, smaller than its predecessor (12B). Despite its size, DALL·E 2 generates 4x better resolution images than DALL·E and it’s preferred by human judges +70% of the time both in caption matching and photorealism.

As they did with DALL·E, OpenAI didn’t release DALL·E 2 (you can always join the never-ending waitlist). However, they open-sourced CLIP which, although only indirectly related to DALL·E, forms the basis of DALL·E 2. (CLIP is also the basis of the apps and notebooks people who can’t access DALL·E 2 are using.) Still, OpenAI’s CEO, Sam Altman, said they’ll eventually release DALL·E models through their API — for now, only a few selected have access to it (they’re opening the model to 1000 people each week).

To continue reading this article, click here.

2 thoughts on “DALL·E 2, Explained: The Promise and Limitations of a Revolutionary AI

  1. Pingback: The Promise and Limitations of a Revolutionary AI « Machine Studying Occasions - The Tech Best

Leave a Reply