Jeff Cui

70 posts

Jeff Cui

Jeff Cui

@jeffacce

PhD CS @ NYU Courant

Katılım Aralık 2014
987 Takip Edilen367 Takipçiler
Sabitlenmiş Tweet
Jeff Cui
Jeff Cui@jeffacce·
We don't need the name of an object to pick it up; we simply need to know where it is and what it looks like. Introducing Contact-Anchored Policies (CAPs): instead of language, we explicitly condition on contacts. Our policy learns object pickup with only 16 hours of data! 🧵
English
5
28
108
13.3K
Jeff Cui retweetledi
Mahi Shafiullah 🏠🤖 RSS 2026 ✈️
Mahi Shafiullah 🏠🤖 RSS 2026 ✈️@notmahi·
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: 🧵
English
11
65
358
92.4K
Jeff Cui retweetledi
Lerrel Pinto
Lerrel Pinto@LerrelPinto·
Turns out you can train humanoid hands without any robot data. The idea in HUG is quite simple: (a) collect human data with smart glasses, (b) train a human manipulation model, (c) retarget to multi-fingered robot hands.
English
4
30
220
32K
Jeff Cui retweetledi
Younghyo Park ✈️RSS2026
Younghyo Park ✈️RSS2026@younghyo_park·
[10/n] Broader implication 2 Learning from human videos or wearables is a promising direction. These paradigms, however, often treat "future state" as its pseudo action label — implicitly assuming perfect tracking, which is effectively a stiff-controller assumption. If our findings generalize, rethinking this assumption could unlock even more of their potential.
Younghyo Park ✈️RSS2026 tweet media
English
1
7
66
33.1K
Jeff Cui retweetledi
Irmak Guzey
Irmak Guzey@irmakkguzey·
Learning from human data requires human-like hardware. Humans use their wrists constantly, but table-top manipulators lack this flexibility. We build upon RUKA and introduce RUKA-v2: a tendon-driven hand with a 2-DOF wrist and finger abduction/adduction 👋✌️
English
7
28
116
9K
Jeff Cui retweetledi
Krish Mehta
Krish Mehta@djkesu1·
World models are neural simulators. But neural simulators need grounding. If you close your eyes and reach out for the coffee cup in front of you, you’ll be able to manipulate it. To pass The Physical Turing Test, we need action loops at scale, irrespective of the modality, and that’s what the bitter lesson teaches us. We are upgrading Simulation 1.0 to 1.5 - generative assets and scenes, and we are calling it PhysReady. [1/]
Palatial@PalatialSim

A child consumes more data in 1 month than any LLM has ever seen. Embodied agents learn by doing, but the data that teaches them is tactile, sensorial and causal. Such data does not exist. To make physical AGI possible, we need to generate this new data at an industrial scale. Enter Palatial: automated infrastructure that converts raw data into sensory rich playgrounds for robots to learn in. Today, we’re unveiling Palatial PhysReady, the first automated sim asset generator (try it ⬇️) [1/5]

English
5
4
42
4.1K
Jeff Cui retweetledi
Siddhant Haldar
Siddhant Haldar@haldar_siddhant·
Robot foundation models are limited by costly real data, while simulation data is plentiful but visually mismatched to reality. We present Point Bridge, a method that enables zero-shot sim-to-real transfer for robot learning with minimal visual alignment. pointbridge3d.github.io
English
4
41
222
19.9K
Jeff Cui retweetledi
Lerrel Pinto
Lerrel Pinto@LerrelPinto·
Introducing YOR. Balancing budget and functionality for a capable mobile robot is always a challenge. To give researchers and hobbyists more options, we built our own open-source one for ~$10k.
English
1
9
54
3.6K
Jeff Cui
Jeff Cui@jeffacce·
@leo_lin6 We've only tested our policy on this gripper, which is open-source and you can build one too! (See hardware section on website.) The iPhone is used for data collection and inference as a wrist camera, giving us RGB, depth, and camera odometry.
English
0
0
1
34
Leo Lin
Leo Lin@leo_lin6·
@jeffacce How hard would it be for people to adapt other grippers to also use this policy? And is the iphone necessary or just for accessibility (or is it VLM related)?
English
1
0
0
87
Jeff Cui
Jeff Cui@jeffacce·
We don't need the name of an object to pick it up; we simply need to know where it is and what it looks like. Introducing Contact-Anchored Policies (CAPs): instead of language, we explicitly condition on contacts. Our policy learns object pickup with only 16 hours of data! 🧵
English
5
28
108
13.3K
Jeff Cui
Jeff Cui@jeffacce·
@_varunnair We train our encoder and policy from scratch! These models use a ResNet-50 backbone for the visual encoder, with a Vector-Quantized Behavior Transformer (VQ-BeT) as the policy head. The model is small enough that it runs at 3Hz on an Intel NUC CPU.
English
0
0
1
39
pfung
pfung@philfung·
@jeffacce saw several other vids from others and it looks super robust. big congrats!
San Francisco, CA 🇺🇸 English
1
0
1
65
Jeff Cui retweetledi
Manan Anjaria
Manan Anjaria@AnjariaManan·
The real gap isn't capability, it's accessibility. We need platforms that labs can actually build, hack and improve without needing Big budgets or NDAs. Something modular, documented, cheap and yet capable enough to conduct hours of research . We present you YOR
Manan Anjaria tweet media
Mahi Shafiullah 🏠🤖 RSS 2026 ✈️@notmahi

Why buy a robot when you can build your own? Meet YOR, our new open-source bimanual mobile manipulator robot – built for researchers and hackers alike for only ~$10k. 🧵👇

English
4
25
135
21K
Jeff Cui retweetledi
Mahi Shafiullah 🏠🤖 RSS 2026 ✈️
Mahi Shafiullah 🏠🤖 RSS 2026 ✈️@notmahi·
Why buy a robot when you can build your own? Meet YOR, our new open-source bimanual mobile manipulator robot – built for researchers and hackers alike for only ~$10k. 🧵👇
English
7
22
167
38.8K
Jeff Cui
Jeff Cui@jeffacce·
The Jetson integration allows us to run our learned policies directly onboard, without having to worry about networking jitter, with multiple RGB streams, base odometry, and proprioception (10x autonomous):
English
1
0
4
129
Jeff Cui
Jeff Cui@jeffacce·
Fully open-source, customizable hardware is the way for robotics research. Introducing Your Own Robot (YOR), a mobile bimanual robot platform for ~$10k.
English
2
4
32
1.5K

Keşfet

@nyuniversity @leo_lin6 @_varunnair @philfung @AnjariaManan @MEnesErciyes @notmahi @elonmusk