🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Modeling the world for action by generating pixel is as wasteful and doomed to failure as the largely-abandoned idea of "analysis by synthesis". Decades ago, there was a big debate in ML about the relative advantages of generative methods vs discriminative methods for

Sharing a fun weekend hack: - closed models (GPT-4, Claude 3) are powerful but untrusted for sensitive inputs - bunch of open LLMs around (Mixtral, Gemma) but not as smart - can we anonymize inputs to GPT-4 w/ a small, open LLM run locally on your MacBook? 🧵some thoughts below:

I absolutely love this type of cognitive-inspired AI research! ❤️

Now imagine a VR powered by interactive diffusion models, where reality is generated, not just rendered!

Lucas Nestler Thanks Lucas!! ideally this would resolve the need for CoT altogether, since the model can "think" about its answer during the denoising process! and also change it if it discovers a better solution

Great essay from Fin Moorhouse and William MacAskill on the Intelligence Explosion quite long, totally worth it forethought.org/research/prepa…

Here's the unedited raw shot from a phone 🤯

Impressive!

Pretty cool way to turn a video model into a policy model.

Shane Gu Yes we are correlation machines with high false discovery rates but with a sufficiently large population it has been working.

Supervised learning has held 3D Vision back for too long. Meet RayZer — a self-supervised 3D model trained with zero 3D labels: ❌ No supervision of camera & geometry ✅ Just RGB images And the wild part? RayZer outperforms supervised methods (as 3D labels from COLMAP is noisy)

He walks like Captain Jack Sparrow lol