Sublime
An inspiration engine for ideas

LAION-400m pretrained weights are now in OpenCLIP https://t.co/51fIQqwMWq. I've done some refactoring to make the modelling function as a stand alone package, pypi release in the works... https://t.co/oLdh8eybFm
LLaVA-3D
A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness
Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the... See more
AKx.comMeta-learning
Jennifer • 4 cards

Extract alpha from multi LLM use. Don't leave anything left on the table. Don't just single-shot, don't stick to one tool. Always use best-of-breed AI all the time. Don't ever settle for less than the best.

StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery
Combines CLIP and StyleGAN to improve text-driven image manipulation.
abs: https://t.co/SPqhRITXbe
code: https://t.co/Lc6R6TodVa https://t.co/lGkFGGlVx3