Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

What if we're approaching robotic manipulation all wrong? Ilya Sutskever makes a fascinating point here: Current approaches throw massive amounts of data at the problem — millions of simulation steps, huge compute. And still, robots can't match human dexterity. Real-world learning of new skills? "Very out of reach." Meanwhile,... humans pick up manipulation tasks almost instantly. Our dexterity is unmatched. And we do it with remarkable robustness — no reward shaping, no curriculum design, no training instability. Why the gap? Evolution. 500M years of optimisation for locomotion, vision, and manipulation. For these ancient sensorimotor tasks, maybe brute-force data isn't the answer. Maybe we should be copying the biological algorithms that already solved it. CC: Dwarkesh Patelshow more

Jack 🤖

13,098 subscribers

31,262 görüntüleme • 11 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Unitree founder Wang Xingxing: In robotics, locomotion and basic motion is mostly solved. But grasping and manipulation—anything related to haptics—hasn’t been solved. That’s the key bottleneck preventing them from being deployed at scale in factories and homes. He says that simulation is much faster for training but for manipulation tasks you still need real-world training data—for now.

Unitree founder Wang Xingxing: In robotics, locomotion and basic motion is mostly solved. But grasping and manipulation—anything related to haptics—hasn’t been solved. That’s the key bottleneck preventing them from being deployed at scale in factories and homes. He says that simulation is much faster for training but for manipulation tasks you still need real-world training data—for now.

Kyle Chan

67,761 görüntüleme • 6 ay önce

2026 is the year of robots: Atlas's hands mimic human dexterity with adaptable grips for various tasks. Tactile sensors enable fine-grain manipulation. Atlas is a step toward the future.

2026 is the year of robots: Atlas's hands mimic human dexterity with adaptable grips for various tasks. Tactile sensors enable fine-grain manipulation. Atlas is a step toward the future.

Chubby♨️

14,750 görüntüleme • 10 ay önce

How to use simulation data for real-world robot manipulation? We present sim-and-real co-training, a simple recipe for manipulation. We demonstrate that sim data can significantly enhance real-world performance, even with notable differences between the sim and the real. (1/n)

How to use simulation data for real-world robot manipulation? We present sim-and-real co-training, a simple recipe for manipulation. We demonstrate that sim data can significantly enhance real-world performance, even with notable differences between the sim and the real. (1/n)

Zhenyu Jiang

45,051 görüntüleme • 1 yıl önce

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

NVIDIA Robotics

13,562 görüntüleme • 9 ay önce

A big part of scaling robot learning to solve real-world problems is that we somehow need to get enough diverse, high-quality data to train our robots to perform useful things. GPT and its fellow large language models were bootstrapped and proved out on a massive dataset of real-world language data. Unfortunately, despite our best efforts, similarly massive datasets don’t really exist for robotics — so, in our unending pursuit of high-quality, useful data, we turn to simulation. I compared a couple recent works on sim-to-real robot manipulation, which discuss how to train perception-driven manipulation policies in simulation, in such a way that they’re useful in the real world. - DextraH-RGB, from NVIDIA - Sim-and-Real Co-Training: A Simple Recipe for Vision-Based Robotic Manipulation, also from NVIDIA — specifically the GEAR lab - Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids, another GEAR lab paper - Local Policies Enable Zero-shot Long-Horizon Manipulation, from CMU (video from DextrAH-RGB)

A big part of scaling robot learning to solve real-world problems is that we somehow need to get enough diverse, high-quality data to train our robots to perform useful things. GPT and its fellow large language models were bootstrapped and proved out on a massive dataset of real-world language data. Unfortunately, despite our best efforts, similarly massive datasets don’t really exist for robotics — so, in our unending pursuit of high-quality, useful data, we turn to simulation. I compared a couple recent works on sim-to-real robot manipulation, which discuss how to train perception-driven manipulation policies in simulation, in such a way that they’re useful in the real world. - DextraH-RGB, from NVIDIA - Sim-and-Real Co-Training: A Simple Recipe for Vision-Based Robotic Manipulation, also from NVIDIA — specifically the GEAR lab - Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids, another GEAR lab paper - Local Policies Enable Zero-shot Long-Horizon Manipulation, from CMU (video from DextrAH-RGB)

Chris Paxton

20,486 görüntüleme • 1 yıl önce

$Ilya Sutskever: We shouldn't just align AI to humans. We should align it to sentient life. That may even be easier, because a sentient AI will empathize with other sentient beings. Dwarkesh Patel: But then humans become a tiny fraction of future sentience. Do we lose control? Ilya Sutskever: It's true. But maybe the answer is to cap the power of the strongest superintelligence.$

Ilya Sutskever: We shouldn't just align AI to humans. We should align it to sentient life. That may even be easier, because a sentient AI will empathize with other sentient beings. Dwarkesh Patel: But then humans become a tiny fraction of future sentience. Do we lose control? Ilya Sutskever: It's true. But maybe the answer is to cap the power of the strongest superintelligence.

vitrupo

53,984 görüntüleme • 11 ay önce

Robots Doing Stuff #61 | Bimanual Nut Pick and Threading This is a real customer task: one step in a longer multi-part sequence we’re training in our labs. Threading requires precise alignment, force control, and fine manipulation. These are foundational capabilities for industrial dexterity applications that are currently inaccessible with traditional automation. We teleoperate these tasks to collect the data that trains our Physical AI. Next stop: autonomous. Visit to learn more about our Physical AI-enabled automation solutions. #Dexterity #DexterousHand #RobotHand #Robotics #RobotsDoingStuff #SanctuaryAI

Robots Doing Stuff #61 | Bimanual Nut Pick and Threading This is a real customer task: one step in a longer multi-part sequence we’re training in our labs. Threading requires precise alignment, force control, and fine manipulation. These are foundational capabilities for industrial dexterity applications that are currently inaccessible with traditional automation. We teleoperate these tasks to collect the data that trains our Physical AI. Next stop: autonomous. Visit to learn more about our Physical AI-enabled automation solutions. #Dexterity #DexterousHand #RobotHand #Robotics #RobotsDoingStuff #SanctuaryAI

Sanctuary AI

38,402 görüntüleme • 4 ay önce

Robots that learn should generalize. Locomotion should show generalization across terrain; manipulation across object categories. Here is a video of “Pick Anything”, demonstrating human like grasps for 66 different objects, ranging in size and shape from air pods to dumbbells to pipe wrenches, using a *single* policy. The system uses the stereo camera head for vision (no wrist cameras) & Sharpa Wave hands for manipulation. It was trained using video and RL (zero teleoperation data). This is research from our Dex-Hands team at Amazon FAR and we will be doing live demos of the system at CoRL 2026 ( Demo 12, Nov. 9 &10). Please come by!

Robots that learn should generalize. Locomotion should show generalization across terrain; manipulation across object categories. Here is a video of “Pick Anything”, demonstrating human like grasps for 66 different objects, ranging in size and shape from air pods to dumbbells to pipe wrenches, using a single policy. The system uses the stereo camera head for vision (no wrist cameras) & Sharpa Wave hands for manipulation. It was trained using video and RL (zero teleoperation data). This is research from our Dex-Hands team at Amazon FAR and we will be doing live demos of the system at CoRL 2026 ( Demo 12, Nov. 9 &10). Please come by!

Jitendra MALIK

27,901 görüntüleme • 1 gün önce

State-of-the-art robot policies often need hundreds of hours of data. What if we needed none? Introducing TiPToP: a manipulation system that zero-shots open-world tasks from pixels and language using vision foundation models and GPU-parallelized Task and Motion Planning (TAMP).

State-of-the-art robot policies often need hundreds of hours of data. What if we needed none? Introducing TiPToP: a manipulation system that zero-shots open-world tasks from pixels and language using vision foundation models and GPU-parallelized Task and Motion Planning (TAMP).

Nishanth Kumar

80,042 görüntüleme • 8 ay önce

🎙️ Excited to introduce one of my favorite projects from the past year: TeleDexter, from the BIGAI dexterity team. It’s a stable, human-level dexterous teleoperation system and a suite of autonomous policies trained with it. Pen spinning, complex in-hand reorientation, and long-horizon tool use—once seen as the holy grail of manipulation—are now unlocked. 🧵👇 The hardware is already here; we have some incredible high-DoF robotic hands. The bottleneck? The controller. Most current systems are stuck in "quasi-static" grasping mode. Meanwhile, dynamic in-hand dexterity has remained severely limited. 🧠 To unlock the massive capabilities of human-like hands, we need to build an excellent "cerebellum" for dexterous hands. TeleDexter solves this with a novel co-tracking approach: it simultaneously tracks both human hand kinematics and object states, beautifully bridging the gap between human intent and robotic control. In order to train a better co-tracking policy that works robustly in the real world, we designed : (1) a hybrid reward design that combines consecutive goal reaching and dense tracking, (2) an action masking strategy during training that enhances sim2real performance, (3) a dexterous curriculum for learning the long-horizon interactions. Each design is inspired by numerous trials and countless real-world experiments. We’ve synthesized all the system details, engineering challenges, and core insights into our latest post. If you're interested in the future of dexterous manipulation, grab a coffee and check it out (9-min read): If you have more time, check out the paper:

🎙️ Excited to introduce one of my favorite projects from the past year: TeleDexter, from the BIGAI dexterity team. It’s a stable, human-level dexterous teleoperation system and a suite of autonomous policies trained with it. Pen spinning, complex in-hand reorientation, and long-horizon tool use—once seen as the holy grail of manipulation—are now unlocked. 🧵👇 The hardware is already here; we have some incredible high-DoF robotic hands. The bottleneck? The controller. Most current systems are stuck in "quasi-static" grasping mode. Meanwhile, dynamic in-hand dexterity has remained severely limited. 🧠 To unlock the massive capabilities of human-like hands, we need to build an excellent "cerebellum" for dexterous hands. TeleDexter solves this with a novel co-tracking approach: it simultaneously tracks both human hand kinematics and object states, beautifully bridging the gap between human intent and robotic control. In order to train a better co-tracking policy that works robustly in the real world, we designed : (1) a hybrid reward design that combines consecutive goal reaching and dense tracking, (2) an action masking strategy during training that enhances sim2real performance, (3) a dexterous curriculum for learning the long-horizon interactions. Each design is inspired by numerous trials and countless real-world experiments. We’ve synthesized all the system details, engineering challenges, and core insights into our latest post. If you're interested in the future of dexterous manipulation, grab a coffee and check it out (9-min read): If you have more time, check out the paper:

Siyuan Huang

12,847 görüntüleme • 3 ay önce

Ilya Sutskever explains the 1% human training loop, and why fully autonomous AI is the wrong target "The thing you really want is for the human teachers that tell, that teach the AI, for them to collaborate with an AI." "You might want to think of it as being in a world where the human teachers do 1% of the world and the work and the AI do 99% of the work." "You don't want it to be 100% AI" "But you do want it to be a human-machine collaboration which teaches the next machine." Most teams are trying to remove the human from the loop. Ilya's design keeps a tiny human control surface and uses it to teach the next generation. The leverage is not zero humans. It is 1% human judgment steering 99% machine work. - Ilya Sutskever (Ilya Sutskever), AI researcher, on Dwarkesh Patel

Ilya Sutskever explains the 1% human training loop, and why fully autonomous AI is the wrong target "The thing you really want is for the human teachers that tell, that teach the AI, for them to collaborate with an AI." "You might want to think of it as being in a world where the human teachers do 1% of the world and the work and the AI do 99% of the work." "You don't want it to be 100% AI" "But you do want it to be a human-machine collaboration which teaches the next machine." Most teams are trying to remove the human from the loop. Ilya's design keeps a tiny human control surface and uses it to teach the next generation. The leverage is not zero humans. It is 1% human judgment steering 99% machine work. - Ilya Sutskever (Ilya Sutskever), AI researcher, on Dwarkesh Patel

Karl Mehta

30,651 görüntüleme • 3 ay önce

Fei-Fei Li on what simulation can do for robotics that real-world data can't: "You play out events that haven't happened, or cannot happen... And while you play it out, you learn how to act in it." "That's really important in robotics, because we cannot possibly have enough real-world data for that." "Here is a real life example: the industry of self-driving cars. Waymo has officially said they use billions of hours of simulation." "Cars are the simplest kind of robots. So clearly simulation plays a huge role in robotic learning." Fei-Fei Li Yunzhu Li World Labs martin_casado

Fei-Fei Li on what simulation can do for robotics that real-world data can't: "You play out events that haven't happened, or cannot happen... And while you play it out, you learn how to act in it." "That's really important in robotics, because we cannot possibly have enough real-world data for that." "Here is a real life example: the industry of self-driving cars. Waymo has officially said they use billions of hours of simulation." "Cars are the simplest kind of robots. So clearly simulation plays a huge role in robotic learning." Fei-Fei Li Yunzhu Li World Labs martin_casado

a16z

35,794 görüntüleme • 3 ay önce

🚨 Jensen Huang says everyone panicked about the AI data when MOST training data was never REAL to begin with. Ilya Sutskever told the industry pre-training was over. "Ilya said, 'We're out of data,' or something like that. 'Pre-training is over,' or something like that," Huang says. "The industry panicked, you know, that this is the end of AI." "And of course, of course that's obviously not true. We're gonna keep on scaling the amount of data that we have to train with." "A lot of that data is probably gonna be synthetic." That's where the panic came from — synthetic data sounds like cheating. "Most of the data that we are training, that we teach each other with, inform each other with, is synthetic." "It's synthetic because it didn't come out of nature." "You created it. I'm consuming it. I modify it, augment it, I regenerate it, somebody else consumes it." The textbook in your hand is synthetic. The post you're reading is synthetic. The lecture you took is synthetic. Nature didn't make any of it. Humans did. AI just learned to do the same thing — faster. "Training is now limited by compute," Huang says. "Data is now limited by compute." The data wall wasn't a wall. It was a mirror. If you're new here, follow @AiEvolutio for the latest on ChatGPT, Claude, and the AI tools shaping how we work and create. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

🚨 Jensen Huang says everyone panicked about the AI data when MOST training data was never REAL to begin with. Ilya Sutskever told the industry pre-training was over. "Ilya said, 'We're out of data,' or something like that. 'Pre-training is over,' or something like that," Huang says. "The industry panicked, you know, that this is the end of AI." "And of course, of course that's obviously not true. We're gonna keep on scaling the amount of data that we have to train with." "A lot of that data is probably gonna be synthetic." That's where the panic came from — synthetic data sounds like cheating. "Most of the data that we are training, that we teach each other with, inform each other with, is synthetic." "It's synthetic because it didn't come out of nature." "You created it. I'm consuming it. I modify it, augment it, I regenerate it, somebody else consumes it." The textbook in your hand is synthetic. The post you're reading is synthetic. The lecture you took is synthetic. Nature didn't make any of it. Humans did. AI just learned to do the same thing — faster. "Training is now limited by compute," Huang says. "Data is now limited by compute." The data wall wasn't a wall. It was a mirror. If you're new here, follow @AiEvolutio for the latest on ChatGPT, Claude, and the AI tools shaping how we work and create. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

AI Evolution

15,565 görüntüleme • 5 ay önce

Robot trained entirely in simulation! 🤯 Sudo AI just released #sudo R1. This could be the moment the data bottleneck in robotics finally breaks, a major step forward for getting simulation-trained robots to actually work in the real world. #sudo R1 is a manipulation foundation model that reliably picks objects it has never seen: glass, fabric, metal, irregular shapes. All with zero real-world training data. No tele-op. No demos. No manual labeling. Pure simulation. And picking isn't just any skill, it's the foundation. Almost everything we want robots to do starts with being able to reliably pick something up. Data collection is the bottleneck holding the whole field back. Every new task means hours of tele-op and armies of labelers. If simulation alone can actually deliver production-grade picking, that bottleneck starts to disappear. Really exciting to see! If the results hold, the path to general-purpose robots just got a whole lot shorter. 👇 Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Robot trained entirely in simulation! 🤯 Sudo AI just released #sudo R1. This could be the moment the data bottleneck in robotics finally breaks, a major step forward for getting simulation-trained robots to actually work in the real world. #sudo R1 is a manipulation foundation model that reliably picks objects it has never seen: glass, fabric, metal, irregular shapes. All with zero real-world training data. No tele-op. No demos. No manual labeling. Pure simulation. And picking isn't just any skill, it's the foundation. Almost everything we want robots to do starts with being able to reliably pick something up. Data collection is the bottleneck holding the whole field back. Every new task means hours of tele-op and armies of labelers. If simulation alone can actually deliver production-grade picking, that bottleneck starts to disappear. Really exciting to see! If the results hold, the path to general-purpose robots just got a whole lot shorter. 👇 Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

25,901 görüntüleme • 6 ay önce

Teaching robots real dexterity has always been a challenge. But what if they could handle tools like a human? DexterityGen (DexGen) is a new system that helps robots use their hands better. It improves how they grip, move, and handle objects… from holding a pen to using a screwdriver. DexGen learns in simulation and refines its skills in the real world, making robotic hands much more useful. What makes DexGen special? ✅ Smarter movements that refine rough actions into precise skills ✅ Trained on a massive collection of dexterous tasks for better learning ✅ Better teleoperation that makes robotic hand control easier and safer ✅ Handles real-world challenges like small objects, tricky angles, and gravity This moves robots closer to real dexterity. It makes tool use more natural, improves stability, and brings robotic hands one step closer to human-level skill. Seen at Zhao-Heng Yin 🫶 Github: Paper:

Teaching robots real dexterity has always been a challenge. But what if they could handle tools like a human? DexterityGen (DexGen) is a new system that helps robots use their hands better. It improves how they grip, move, and handle objects… from holding a pen to using a screwdriver. DexGen learns in simulation and refines its skills in the real world, making robotic hands much more useful. What makes DexGen special? ✅ Smarter movements that refine rough actions into precise skills ✅ Trained on a massive collection of dexterous tasks for better learning ✅ Better teleoperation that makes robotic hand control easier and safer ✅ Handles real-world challenges like small objects, tricky angles, and gravity This moves robots closer to real dexterity. It makes tool use more natural, improves stability, and brings robotic hands one step closer to human-level skill. Seen at Zhao-Heng Yin 🫶 Github: Paper:

Ilir Aliu

51,535 görüntüleme • 1 yıl önce

Today, we publicly released RoboCasa365, a large-scale simulation benchmark for training and systematically evaluating generalist robot models. Built upon our original RoboCasa framework, it offers: • 2,500 realistic kitchen environments; • 365 everyday tasks (basic skills + long-horizon mobile manipulation); • Over 3,200 objects with many articulated fixtures/appliances. All are designed for fully controlled, reproducible benchmarking of robotic policies. Progress in robotic foundation models is real. But it’s still hard to answer basic questions like: How close are we to general-purpose autonomy? What factors drive generalization? What are the model/data scaling curves like? Real-world eval is slow and noisy, and existing sims (like LIBERO, which we built 3 years ago) often lack sufficient task and scene diversity. This benchmark comes with 2,200+ hours of demonstrations and 500K+ trajectories to support studies of multi-task training, pretraining, and continual learning at scale. Check it out at

Today, we publicly released RoboCasa365, a large-scale simulation benchmark for training and systematically evaluating generalist robot models. Built upon our original RoboCasa framework, it offers: • 2,500 realistic kitchen environments; • 365 everyday tasks (basic skills + long-horizon mobile manipulation); • Over 3,200 objects with many articulated fixtures/appliances. All are designed for fully controlled, reproducible benchmarking of robotic policies. Progress in robotic foundation models is real. But it’s still hard to answer basic questions like: How close are we to general-purpose autonomy? What factors drive generalization? What are the model/data scaling curves like? Real-world eval is slow and noisy, and existing sims (like LIBERO, which we built 3 years ago) often lack sufficient task and scene diversity. This benchmark comes with 2,200+ hours of demonstrations and 500K+ trajectories to support studies of multi-task training, pretraining, and continual learning at scale. Check it out at

Yuke Zhu

24,748 görüntüleme • 8 ay önce

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork

Pascal Bornet

186,921 görüntüleme • 7 ay önce

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

MTS

18,002 görüntüleme • 2 ay önce