Image Models
topic on 3 shows · 4 statements across 4 episodes
No Priors 
the a16z Podcast 
Big Technology
4 statements about Image Models, every show
Stern: The 'Hamster Test' Is Essential for Evaluating Image Models
“Now, my current test for all these models, image models, is counting hamsters, and it has gotten better. It still may make the mistake from time to time, but if you're not doing the hamster test, you're not really testing models.”
Sherman Wu: Smaller size of image models drives faster iteration and proliferation
“Image models tend to be way smaller and like you can iterate on it a lot faster. Like that's why you get that crazy cool proliferation of like the image model side.”
Doshi: AI Image Models Homogenize Styles by Overfitting to Average Human Preference
“I think that the models are probably too curated and maybe overfit to be based on human preference. And human preference isn't your human preference. It's your preference. I mean, it's some kind of average of human preference.”
Parikh: Generative video progress will lag behind LLM and image advancements
“I think that might be harder in video and I wonder if there is something that we are kind of fundamentally missing in terms of how we approach video generation. So it's not quite answering what you asked me, but I do think that it might be a little bit slower …”