aichannel

is there any interest in cast summaries like these of new ML papers from big companies? 

Apple’s approach: https://arxiv.org/abs/2412.06329 [2412.06329] Normalizing Flows are Capable Generative Models

Meta’s approach: https://arxiv.org/abs/2210.02747 [2210.02747] Flow Matching for Generative Modeling

yes but dumbed down for the retards like me

CEO of Mid Curve Capital. Building @daosdotworld

yeah ever since then modifications or alternate angles on that have been the approach most nerds have researched but this one is saying hey transformers from LLMs with really basic architectures can kinda match and sometimes beat really really complicated diffusion models

I do know a bit about stable diffusion ..

multiple approaches to doing image stuff, most recently the hype has been one way for images one way for LLMs, apple says hey the LLM approach can be good for images too