For a split second there I believed there is a new Distill publication! Their articles were the most inspirational and eye-opening resource on my beginnings of ML journey, the quality of visualizations definitely made lasting impact on my mental models of _what is going on_.
My favourite goes definitely to The Building Blocks of Interpretability (https://distill.pub/2018/building-blocks/), those images landed in a lot of my university presentations and the dog made everyone immediately interested ;-)
rahen 6 hours ago [-]
There is ongoing research on neural cellular automata, as they seem to be a very efficient way to generate pretraining tokens: https://arxiv.org/html/2603.10055v1
ticulatedspline 3 hours ago [-]
Morphogenesis data compression is pretty impressive, the human genome is only ~700Mb of data.
api 2 hours ago [-]
Ahh but it requires the laws of physics as a runtime.
56 minutes ago [-]
smusamashah 6 hours ago [-]
This has been shared a few times here. It looks cool but I don't understand the usefulness of this tech. This is likely a demo to show some capability, what that is though I don't understand.
pulkitsh1234 5 hours ago [-]
well it attempts to explain how individual cells can combine to create complex creatures (like a lizard) AND how the creature can have features like healing, regeneration, etc.
Rendered at 02:50:21 GMT+0000 (Coordinated Universal Time) with Vercel.
My favourite goes definitely to The Building Blocks of Interpretability (https://distill.pub/2018/building-blocks/), those images landed in a lot of my university presentations and the dog made everyone immediately interested ;-)