Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Robotic insight 2: One undocumented feature of VLA models is that they can generalize pretty well from very little data, if you run them on good hardware. Here, with 35 demonstrations (5 mins in total), robot can pick up small pcb from any position and at any rotation angle.

ihorbeaver's profile picture

Igor Kulakov (MicroFactory)

8,971 subscribers

47,571 Aufrufe • vor 7 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Over the last few months, we’ve been thinking about how to learn from “off-domain” data - data from non-robot sources like video or simulation. These data sources are not quite good enough to learn policies (even monolithic VLA models) directly, but they still contain lots of information that can be useful for generalizable robot control. How can we develop robot learning models that are able to make use of this type of data for generalizable control? In new work, that we call HAMSTER, we show that VLMs can be useful for enabling robotic learning from off-domain data, but specifically when used through hierarchical VLA architectures. We show that this class of models can learn generalizable robot policies for the real world from large-scale, off-domain data. A 🧵 (1/10)
1:00

Over the last few months, we’ve been thinking about how to learn from “off-domain” data - data from non-robot sources like video or simulation. These data sources are not quite good enough to learn policies (even monolithic VLA models) directly, but they still contain lots of information that can be useful for generalizable robot control. How can we develop robot learning models that are able to make use of this type of data for generalizable control? In new work, that we call HAMSTER, we show that VLMs can be useful for enabling robotic learning from off-domain data, but specifically when used through hierarchical VLA architectures. We show that this class of models can learn generalizable robot policies for the real world from large-scale, off-domain data. A 🧵 (1/10)

Abhishek Gupta

11,994 Aufrufe • vor 1 Jahr

Andre Agassi on Novak Djokovic 🐐: “He is like an amoeba that just that just swallows you and drags you under… It’s like he want to beat your mom too.. He can hurt you from any part of the court in a defensive position. He can hurt you from any part of the court in an offensive position”
0:27

Andre Agassi on Novak Djokovic 🐐: “He is like an amoeba that just that just swallows you and drags you under… It’s like he want to beat your mom too.. He can hurt you from any part of the court in a defensive position. He can hurt you from any part of the court in an offensive position”

Nicolas Ally ✝️

135,600 Aufrufe • vor 11 Monaten

The First 100 begins now. Verify Quality is live on PrismaX. For the first time, anyone can score the robot training data that models learn from, earn points, and compete to become one of The First 100. Better data. Better models. The standard starts with you.
0:34

The First 100 begins now. Verify Quality is live on PrismaX. For the first time, anyone can score the robot training data that models learn from, earn points, and compete to become one of The First 100. Better data. Better models. The standard starts with you.

PrismaX

221,936 Aufrufe • vor 2 Monaten

Here is one of the best answers I have heard from a vet on if they gave any advice to a young player who is playing in their first playoff game. “No none at all” Adams goes on to say that “some lessons you have to learn by going through them” He also said that the vets will be there to pick them up if they fall.
0:43

Here is one of the best answers I have heard from a vet on if they gave any advice to a young player who is playing in their first playoff game. “No none at all” Adams goes on to say that “some lessons you have to learn by going through them” He also said that the vets will be there to pick them up if they fall.

Lachard Binkley

99,650 Aufrufe • vor 1 Jahr

What if one AI brain could run every robot on the planet—from a humanoid to a warehouse arm—all at once? 🧠 , CEO and Co-Founder, and Abhinav Gupta, President and Co-Founder of Skild AI, explain how they are building "OmniBrain," a universal foundation model designed to generalize intelligence across any robot form factor and task. 📺 Watch the episode:
0:28

What if one AI brain could run every robot on the planet—from a humanoid to a warehouse arm—all at once? 🧠 , CEO and Co-Founder, and Abhinav Gupta, President and Co-Founder of Skild AI, explain how they are building "OmniBrain," a universal foundation model designed to generalize intelligence across any robot form factor and task. 📺 Watch the episode:

NVIDIA Robotics

29,512 Aufrufe • vor 4 Monaten

In Avowed You’re not locked to any specific class. You can have 2 load outs at any given time, so mix and match types for different situations. Press Y to Change Load outs on the go. Very cool feature in an RPG. From XboxSeriesX
0:52

In Avowed You’re not locked to any specific class. You can have 2 load outs at any given time, so mix and match types for different situations. Press Y to Change Load outs on the go. Very cool feature in an RPG. From XboxSeriesX

Jamie Moran

144,658 Aufrufe • vor 1 Jahr

Did you know that in Blender you can easily make the rotation of one object drive the rotation of another? ✨ 1. Select your object 2. Hover over one of its rotation fields and press CTRL+D 3. In the popup select the rotation of another object 4. You can write "smoothstep(0, tau, var) * 2" to both limit the rotation and make it non-linear. Tau is 180 degrees, and it's the end angle for the wheel, and 2 is the end angle for the gate leaf in radians (3 is approx. 180, so 2 is approx 2/3 of that!) 5. To copy the animation to the second leaf, select both, press CTRL+L, and choose 'Link Animation Data'. To invert the angle, add a minus sign, and you're done! 👌 #b3d
0:53

Did you know that in Blender you can easily make the rotation of one object drive the rotation of another? ✨ 1. Select your object 2. Hover over one of its rotation fields and press CTRL+D 3. In the popup select the rotation of another object 4. You can write "smoothstep(0, tau, var) * 2" to both limit the rotation and make it non-linear. Tau is 180 degrees, and it's the end angle for the wheel, and 2 is the end angle for the gate leaf in radians (3 is approx. 180, so 2 is approx 2/3 of that!) 5. To copy the animation to the second leaf, select both, press CTRL+L, and choose 'Link Animation Data'. To invert the angle, add a minus sign, and you're done! 👌 #b3d

passivestar

107,167 Aufrufe • vor 3 Jahren

This robot assistant from the NVIDIA CES Keynote on Monday is going viral. Nader Khalil🍊 explains all the hottest emerging AI trends in one demo: AI applications in 2026 will be multi-model, multi-modal, hybrid cloud/local, use open source models as well as proprietary models, control robots and embedded devices in the physical world, and have voice interfaces. (And the demo had a cute robot *and* a cute dog. Gold.) The demo was built with Pipecat AI. NVIDIA posted a really nice technical walk-through and complete code. The Reachy Mini robot from Hugging Face is open source hardware. (You can order it now, I have one!). You can run the assistant locally on your own hardware, in the cloud, or both.
2:26

This robot assistant from the NVIDIA CES Keynote on Monday is going viral. Nader Khalil🍊 explains all the hottest emerging AI trends in one demo: AI applications in 2026 will be multi-model, multi-modal, hybrid cloud/local, use open source models as well as proprietary models, control robots and embedded devices in the physical world, and have voice interfaces. (And the demo had a cute robot *and* a cute dog. Gold.) The demo was built with Pipecat AI. NVIDIA posted a really nice technical walk-through and complete code. The Reachy Mini robot from Hugging Face is open source hardware. (You can order it now, I have one!). You can run the assistant locally on your own hardware, in the cloud, or both.

kwindla

49,028 Aufrufe • vor 7 Monaten

Quadrupeds are fast. Agile. Great at locomotion. But can they manipulate? A new approach from Carnegie Mellon University, Google DeepMind, and Bosch is teaching quadrupedal robots to do more than walk, they’re learning to interact. It’s called Human2LocoMan: a system that uses human data to pretrain robot policies before finetuning on real hardware. The result? A four-legged robot that can walk, carry, organize, scoop, and sort with both single and dual-arm control. By pretraining on human motion, they cut the amount of robot data in half—while improving success rates by over 80% in unfamiliar environments. Their Modularized Cross-Embodiment Transformer (MXT) learns from both human and robot demonstrations, then generalizes those skills to physical tasks—no hardcoded behaviors required. It’s locomotion and manipulation. A quadruped that can walk and clean up after itself?
1:21

Quadrupeds are fast. Agile. Great at locomotion. But can they manipulate? A new approach from Carnegie Mellon University, Google DeepMind, and Bosch is teaching quadrupedal robots to do more than walk, they’re learning to interact. It’s called Human2LocoMan: a system that uses human data to pretrain robot policies before finetuning on real hardware. The result? A four-legged robot that can walk, carry, organize, scoop, and sort with both single and dual-arm control. By pretraining on human motion, they cut the amount of robot data in half—while improving success rates by over 80% in unfamiliar environments. Their Modularized Cross-Embodiment Transformer (MXT) learns from both human and robot demonstrations, then generalizes those skills to physical tasks—no hardcoded behaviors required. It’s locomotion and manipulation. A quadruped that can walk and clean up after itself?

Lukas Ziegler

61,610 Aufrufe • vor 1 Jahr

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:
0:29

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

42,864 Aufrufe • vor 6 Monaten

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)
0:44

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)

Abhishek Gupta

34,775 Aufrufe • vor 1 Monat

President Museveni: We don't have any problem with refugees, when they come, we give them security and we don't also insist on putting them in the refugee camps. If they can move out and do business, it's okay. If you go to town here, it’s full of people from Somalia, Ethiopia, Eritrea, Congo. This is part of our ideology #ChimpReportsNews
5:46

President Museveni: We don't have any problem with refugees, when they come, we give them security and we don't also insist on putting them in the refugee camps. If they can move out and do business, it's okay. If you go to town here, it’s full of people from Somalia, Ethiopia, Eritrea, Congo. This is part of our ideology #ChimpReportsNews

ChimpReports

58,382 Aufrufe • vor 1 Jahr

If you have any young person in your life I would please ask you (when you get a free minute) to listen to this interview and share it with them, such valuable insight, great advice from sharing a personal journey and an understanding of why that is mature beyond any of our years. “Girls Play On” Jemima Montag
1:40

If you have any young person in your life I would please ask you (when you get a free minute) to listen to this interview and share it with them, such valuable insight, great advice from sharing a personal journey and an understanding of why that is mature beyond any of our years. “Girls Play On” Jemima Montag

Ray Boyne

79,795 Aufrufe • vor 2 Jahren

THIS DEVELOPER LINES UP EVERY SMALL COMPUTER FROM RASPBERRY PI TO MAC MINI - AND EVERY SINGLE ONE CAN RUN LOCAL AI Raspberry Pi Zero 2W for $15, Odroid, thin client, AMD BC250 - a whole range of computers from credit card size to full desktop replacement AMD BC250 is an entire mid-range gaming computer on one card - they come in boxes of 12 and every single one runs local LLMs Mac Mini, N97 with 16GB RAM, 13th gen i7 - any of them replaces $236/month in subscriptions and costs $3/month in electricity you don't need an expensive server - even a $15 Raspberry Pi can run something useful locally without cloud and without subscriptions $600 once, one evening of setup - and AI that runs 24/7 on hardware that might already be sitting in your house
2:30

THIS DEVELOPER LINES UP EVERY SMALL COMPUTER FROM RASPBERRY PI TO MAC MINI - AND EVERY SINGLE ONE CAN RUN LOCAL AI Raspberry Pi Zero 2W for $15, Odroid, thin client, AMD BC250 - a whole range of computers from credit card size to full desktop replacement AMD BC250 is an entire mid-range gaming computer on one card - they come in boxes of 12 and every single one runs local LLMs Mac Mini, N97 with 16GB RAM, 13th gen i7 - any of them replaces $236/month in subscriptions and costs $3/month in electricity you don't need an expensive server - even a $15 Raspberry Pi can run something useful locally without cloud and without subscriptions $600 once, one evening of setup - and AI that runs 24/7 on hardware that might already be sitting in your house

Cortex

20,255 Aufrufe • vor 1 Monat