Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

zero shot towel folding attempt. RL next

atharva ☆

12,134 subscribers

20,459 views • 2 months ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

17 Comments

atharva ☆2 months ago

📍

Lucas2 months ago

you should train a reward model on how smooth or elegant a motion is, pairwise with a bunch of uni or teleop samples. GRPO with that comes out so much better!

Jim Bohnslav2 months ago

info on the hardware?

Chidhambararajan R (a.k.a Chidha)2 months ago

Are those 2 s0101 arms?

OT2 months ago

BIG YAM spotted

별별이2 months ago

lets go higher!!

abhi2 months ago

wowza

Andrew I. Christianson2 months ago

this is good stuff 🔥🔥🔥

Mr. Rc2 months ago

woahhhhhhhhhhhhhhhhhhh

Sibhi Balamurugan2 months ago

Waww

Tirth Gada2 months ago

lessgoo

Dev2 months ago

cooked 🫡

Sahil Apage2 months ago

lessgo finally!!

Rishikesh2 months ago

Insane, are you in Bangalore?

ishaan2 months ago

arm kit? do share any blog you write on this newbie here starting again after a long time :)

sarayu2 months ago

my goat

4nshh2 months ago

Finallyyyy

Related Videos

Oops, towel folding i2RT

Oops, towel folding i2RT

ETH Robotics Club

25,238 views • 11 months ago

A reward model that works, zero-shot, across robots, tasks, and scenes? Introducing Robometer: Scaling general-purpose robotic reward models with 1M+ trajectories. Enables zero-shot: online/offline/model-based RL, data retrieval + IL, automatic failure detection, and more! 🧵 (1/12)

A reward model that works, zero-shot, across robots, tasks, and scenes? Introducing Robometer: Scaling general-purpose robotic reward models with 1M+ trajectories. Enables zero-shot: online/offline/model-based RL, data retrieval + IL, automatic failure detection, and more! 🧵 (1/12)

Jesse Zhang ✈️ CoRL 2026

103,628 views • 8 months ago

connor has him folding under zero pressure

connor has him folding under zero pressure

ilya bizanov

134,066 views • 7 months ago

Midnight will help stop Kelp DAO hacks with next gen private DEXs, Trusted Execution Environments, Multi-Party Computation, and folding zero-knowledge proofs.

Midnight will help stop Kelp DAO hacks with next gen private DEXs, Trusted Execution Environments, Multi-Party Computation, and folding zero-knowledge proofs.

Input Output Group

18,160 views • 6 months ago

jeongin was done with his towel so he was folding it really neatly and cutely 🥹🫳

jeongin was done with his towel so he was folding it really neatly and cutely 🥹🫳

lena ✰

22,986 views • 1 year ago

Ready for the next RL

Ready for the next RL

Rolling Loud

241,245 views • 1 year ago

RL from Next talks about Brandy

RL from Next talks about Brandy

Dionne

12,947 views • 9 months ago

Meet BFM-Zero: A Promptable Humanoid Behavioral Foundation Model w/ Unsupervised RL👉 🧩ONE latent space for ALL tasks ⚡Zero-shot goal reaching, tracking, and reward optimization (any reward at test time), from ONE policy 🤖Natural recovery & transition

Meet BFM-Zero: A Promptable Humanoid Behavioral Foundation Model w/ Unsupervised RL👉 🧩ONE latent space for ALL tasks ⚡Zero-shot goal reaching, tracking, and reward optimization (any reward at test time), from ONE policy 🤖Natural recovery & transition

Yitang Li

84,263 views • 1 year ago

New video of Figure's Helix AI model in action: a human keeps tossing a towel in the middle, the humanoid robot keeps folding away.

New video of Figure's Helix AI model in action: a human keeps tossing a towel in the middle, the humanoid robot keeps folding away.

The Humanoid Hub

115,642 views • 1 year ago

2pac ft RL of Next - Until The End Of Time

2pac ft RL of Next - Until The End Of Time

HomeOfHipHop

21,018 views • 2 months ago

RL from Next Perform 'Too Close' on The Cadillac Chronicles

RL from Next Perform 'Too Close' on The Cadillac Chronicles

RNB RADAR

566,476 views • 1 year ago

Which Raptor threw a towel at Josh after he shot em up?

Which Raptor threw a towel at Josh after he shot em up?

New York Basketball

993,758 views • 9 months ago

Shai flopped on every single shot attempt.

Shai flopped on every single shot attempt.

House of Lowlights

23,113,436 views • 5 months ago

"NEXT DOESN'T HAVE THE DISCOGRAPHY FOR A VERZUZ." - RL?‼️🤔

"NEXT DOESN'T HAVE THE DISCOGRAPHY FOR A VERZUZ." - RL?‼️🤔

SoulFood66

65,156 views • 20 days ago

Boston Dynamics taught Atlas a Rabona kick. Learned from human mocap, retargeted to Atlas, trained in sim via RL, deployed zero-shot to the real robot. Soccer skills demand whole-body coordination, and similar recipe transfers to warehouse work.

Boston Dynamics taught Atlas a Rabona kick. Learned from human mocap, retargeted to Atlas, trained in sim via RL, deployed zero-shot to the real robot. Soccer skills demand whole-body coordination, and similar recipe transfers to warehouse work.

The Humanoid Hub

11,338 views • 5 months ago

Our model can now learn from its own experience with RL! Our new π*0.6 model can more than double throughput over a base model trained without RL, and can perform real-world tasks: making espresso drinks, folding diverse laundry, and assembling boxes. More in the thread below.

Our model can now learn from its own experience with RL! Our new π*0.6 model can more than double throughput over a base model trained without RL, and can perform real-world tasks: making espresso drinks, folding diverse laundry, and assembling boxes. More in the thread below.

Physical Intelligence

713,427 views • 1 year ago

Introducing RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning. 7 real robot tasks, 900/900 successes. Up to 250 consecutive trials in one task, running 2 hours nonstop without failure. High success rate against physical disturbances, zero-shot, and few-shot adaptation Our first step toward a deployable robot learning system.

Introducing RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning. 7 real robot tasks, 900/900 successes. Up to 250 consecutive trials in one task, running 2 hours nonstop without failure. High success rate against physical disturbances, zero-shot, and few-shot adaptation Our first step toward a deployable robot learning system.

Kun Lei

92,978 views • 1 year ago

Every Clutch Shot Attempt From Arkansas Darius Acuff Jr.

Every Clutch Shot Attempt From Arkansas Darius Acuff Jr.

NBA Replays

122,930 views • 4 months ago

Which shop you hitting if you attempt this shot?

Which shop you hitting if you attempt this shot?

Stingr Golf

80,626 views • 7 months ago

Every Jimmy Butler Game Winning/Game Tying Shot Attempt

Every Jimmy Butler Game Winning/Game Tying Shot Attempt

NBA Replays

111,082 views • 5 months ago