Improving Text-to-Image Models by Prompt Engineering with RLHF

Text-to-image models like Stable Diffusion, DALL-E 2, and Midjourney can generate impressive, artistic images — if you’re good at prompt engineering.
However, many users simply want to generate aesthetically pleasing images to include in blog posts, presentations, and other materials. So, what about the rest of us who haven’t had the opportunity to learn all the magical keywords, such as “trending on Artstation,” and understand how they function in various contexts?

The …

blog dall dall-e dall-e 2 diffusion engineering generated generative-model good human feedback image images materials midjourney presentations prompt rest rlhf stable diffusion text text-to-image