
Mahi Shafiullah 🏠🤖
@notmahi • 3,355 subscribers
Trying to understand the emergence of generally intelligent robotic behavior at @berkeley_ai. Previously @CILVRatNYU @MIT & @Apple AI/ML fellow.
Shorts
Videos

Robots are the bottleneck in scaling robotics, and learning from human video promises to solve it. But how can chaotic human data ever measure up to sanitized, lab-made teleoperation data? Introducing Do as I Do: establishing a much needed correspondence between human videos and dexterous robot data. Some fun insights below: 🧵
Mahi Shafiullah 🏠🤖95,244 次观看 • 2 个月前

Best ideas are often the simplest in hindsight. Meet Contact-Anchored Policies (CAP)🧢: by conditioning policies on physical contact (vs language) we achieve env & embodiment generalization with super low resources. This policy ⬇️ learned to pick from scratch w/ 16 hrs of data 🧵
Mahi Shafiullah 🏠🤖34,528 次观看 • 6 个月前

I'm at #RSS2026 – presenting MolmoSpaces on Tuesday & CAP (Contact Anchored Policies) on Wednesday. I've been thinking a lot about what robotics 2-5 years from now looks like: beyond teleop and position control. If you're interested about anything from robot free/human data or sim evals to force/torque controlled dexterous hands, let's chat! 🧵
Mahi Shafiullah 🏠🤖11,649 次观看 • 1 个月前

Wouldn’t it be nice if you could bring a robot home, give it a video of your room, and immediately start asking it to move objects around? Turns out, now you can! Introducing OK-Robot, a zero-shot language-specified pick & drop system that we built with exactly ZERO training! 🧵
Mahi Shafiullah 🏠🤖56,630 次观看 • 2 年前

If you want robots that can just live with you & help 24/7, it needs to build & update its memory on the fly. Current semantic memory representations like VoxelMap from OK-Robot can't change with the world. That's why we built DynaMem: dynamic memory for a changing, open world!
Mahi Shafiullah 🏠🤖29,706 次观看 • 1 年前
没有更多内容可加载