How to learn stuff outside of LLMs? I talked to my friends recently that told me they have their own image generation model, that's architecture agnostic and can work with Midjourney, StableDiffusion etc. But works on their GPUs. I have no idea what they're talking about. How can I start digging into the topic? I'd love to understand how startups like photoai.com or headshotpro.com managed to build their... Source: 8 months ago

The idea is to code up a website that will let brothers upload 10-20 of their pictures, and using AI will professionally generate AI headshots (like headshotpro.com quality), and put them all together in a nice composite layout ready for ordering prints. It would definitely be cheaper than hiring a photographer or an online service to come and handle it. Source: about 1 year ago

Many settings affect the outputs in interesting ways, but that's half the fun. These LoRAs are very lightly trained; more training may or may not help. The competitions are also performed using zero-shot text guessing, and if Facebook said it, you can bet that's actually Meta AI saying it, and they are leaders in the field. Source: over 1 year ago

You should look at the entire niche of MAE-related papers, that's quite exciting, and the neuroscience-inspired stream of stuff like Barlow Twins. As well, the official Facebook AI blog is surprisingly good coverage of much of the interesting un/semi-supervised DL research FAIR does, and worth going through. Source: almost 2 years ago