Description

Saudi Arabia's $40 billion investment in AI, making them the largest investor in the field.

OpenAI's CEO Sam Altman's prediction that compute will be the currency of the future.

Vid2Robot, a video-based learning framework that directly produces robot actions given a video demonstration of a manipulation task and the robot's current visual observations.

TexDreamer, the first zero-shot multimodal high-fidelity 3D human texture generation model.

Contact:  sergi@earkind.com

Timestamps:

00:34 Introduction

01:50 Saudi Arabia Plans $40 Billion Push Into Artificial Intelligence

03:14 Sam Altman on GPT-5 and the future of AI

04:46 Price Per Part in Lego Sets

05:57 Fake sponsor

07:52 Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformers

09:35 mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding

11:31 TexDreamer: Towards Zero-Shot High-Fidelity 3D Human Texture Generation

13:20 Outro