Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Practice makes perfect. A truly agentic robot shouldn’t just know what to do. It should also figure out how to get better. What if a robot could watch a demonstration, figure out what it needs to practice, and learn how to get better through practice? Introducing RPG: Guided Self-Improvement...

wangyenjen's profile picture

Yen-Jen Wang

1,093 subscribers

160,963 Aufrufe • vor 2 Tagen •via X (Twitter)

BildungGesundheit & WellnessWissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

8 Kommentare

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

What should the robot practice? RPG uses real-world demonstrations to reconstruct simplified simulation tasks that target the capabilities needed for deployment. These tasks need not exactly reproduce the real-world scene. They need to provide useful practice for learning the required skills.

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

How should the robot improve through practice? RPG uses execution feedback, privileged simulator state, and a video analyzer to diagnose failures. Demonstrations provide a reference for guiding improvement. The agent adds and revises reusable skills, then tests the changes across tasks before retaining them in a shared skill library.

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

15 rounds of practice. 28.6% → 95.0%. Across 22 manipulation tasks: • RPG (Gemini 3.8 Flash): 95.0% • ASPIRE: 75.5% • GPT-6 Astra Pro + CaP-Agent0: 60.0% • Gemini + CaP-Agent0: 48.6% • RATs: 41.8% RPG gets there by improving the execution system and shared skills — without updating the foundation model weights. The model stays the same. The agent system gets better.

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

The real test is whether the improvements learned in simulation transfer back to the real world. After practice, we freeze the improved system and deploy it on the physical robot. RPG achieves 30/30 successful real-world trials across: 🗄️ Store ball in drawer and close it 🧣 towel folding 🥣 Transfer bowl from left to right The improved system also transfers to additional real-world manipulation tasks. Practice in sim. Improve the system. Go real.

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

So what actually changes as RPG practices? Over 15 rounds, the shared skill library grows from 15 → 38 skills, with: ➕ 23 new skills 🔧 66 revisions Failures become feedback. Feedback becomes reusable skills. Reusable skills help future tasks. Our takeaway: You don’t need a perfect digital twin to get useful sim-to-real improvement. Even with simplified reconstructed environments and practice tasks, coding agents can learn execution improvements that transfer surprisingly well to the real world. Huge thanks to all of our collaborators and co-authors who made this possible — @erichzjiang, @DengShuying, @HaoruXue, @Weirui_Ye, @rocky_duan, @nhaghtal, Shankar Sastry, @pabbeel, and @HaozhiQ! 🌐 Project: 📄 Paper: Also checkout concurrent work SimEX ( from friends in Amazon FAR!

Profilbild von Hershal Rao
Hershal Raovor 2 Tagen

robot practices for 15 rounds and still beats my benchmark scores smh

Profilbild von Yen-Jen Wang
Yen-Jen Wangvor 2 Tagen

🤣

Profilbild von Karan Jagtiani
Karan Jagtianivor 2 Tagen

Picking what to practice is the underrated half. Retrying is cheap for any agent. Knowing which failure is worth a hundred more attempts, and which one was noise, is where the learning actually comes from.

Ähnliche Videos

Robots are getting smarter, but most still fail the same way. They don’t learn from their own mistakes. A new paper proposes something different: a way for robots to self-improve directly from their failures in the real world. It’s called PLD (Probe, Learn, Distill). The idea: instead of collecting endless human demos, let the robot figure out where it fails, learn how to recover, and then distill that knowledge back into its main model. Key takeaways from the research: ✅ Uses residual reinforcement learning to recover from policy failures ✅ Achieves 99% success on LIBERO and 100% on real Franka and YAM arms ✅ Runs hour-long manipulation tasks without human resets ✅ Builds a feedback loop between real-world data and model adaptation Unlike supervised fine-tuning, which relies on humans, PLD learns from the robot’s own experience. By training on its own failure distribution, the model becomes both more efficient and more aligned with the real world. It’s not just a technical shift, it’s a step toward robots that improve themselves through real-world practice. Thanks for sharing, Wenli Xiao !! Paper and demos: —- Weekly robotics and AI insights. Subscribe free:
0:47

Robots are getting smarter, but most still fail the same way. They don’t learn from their own mistakes. A new paper proposes something different: a way for robots to self-improve directly from their failures in the real world. It’s called PLD (Probe, Learn, Distill). The idea: instead of collecting endless human demos, let the robot figure out where it fails, learn how to recover, and then distill that knowledge back into its main model. Key takeaways from the research: ✅ Uses residual reinforcement learning to recover from policy failures ✅ Achieves 99% success on LIBERO and 100% on real Franka and YAM arms ✅ Runs hour-long manipulation tasks without human resets ✅ Builds a feedback loop between real-world data and model adaptation Unlike supervised fine-tuning, which relies on humans, PLD learns from the robot’s own experience. By training on its own failure distribution, the model becomes both more efficient and more aligned with the real world. It’s not just a technical shift, it’s a step toward robots that improve themselves through real-world practice. Thanks for sharing, Wenli Xiao !! Paper and demos: —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

22,891 Aufrufe • vor 11 Monaten

What if robots could improve themselves by learning from their own failures in the real-world? Introducing 𝗣𝗟𝗗 (𝗣𝗿𝗼𝗯𝗲, 𝗟𝗲𝗮𝗿𝗻, 𝗗𝗶𝘀𝘁𝗶𝗹𝗹) — a recipe that enables Vision-Language-Action (VLA) models to self-improve for high-precision manipulation tasks. PLD couples real-world residual reinforcement learning with standard supervised fine-tuning — letting robots discover, recover, and distill their own data flywheel. Quick 🧵
0:59

What if robots could improve themselves by learning from their own failures in the real-world? Introducing 𝗣𝗟𝗗 (𝗣𝗿𝗼𝗯𝗲, 𝗟𝗲𝗮𝗿𝗻, 𝗗𝗶𝘀𝘁𝗶𝗹𝗹) — a recipe that enables Vision-Language-Action (VLA) models to self-improve for high-precision manipulation tasks. PLD couples real-world residual reinforcement learning with standard supervised fine-tuning — letting robots discover, recover, and distill their own data flywheel. Quick 🧵

Wenli Xiao

186,079 Aufrufe • vor 11 Monaten

15–30 minutes of real-world robot data. That's now enough to go from sim-to-real failure to working robot. Let’s see… You train a robot in simulation. You deploy it in the real world. It fails. The physics don't match. So you try to fine-tune it with real data, but… you never have enough real data, and the fine-tuning breaks everything the simulation taught it. SimDist fixes this with one key decision: don't transfer the policy. Transfer the world model. Keep the reward and value knowledge from simulation frozen. Only update the part that's actually wrong, how the robot predicts physics. Now the robot doesn't have to relearn the entire task in the real world. It already knows what success looks like. It just needs to correct its understanding of how the real world moves. The part that makes this work: they also trained on failures and recoveries; not just perfect demonstrations. Without that, the planner finds the gaps and exploits them. With it, the robot can tell a good future from a bad one. That's all it needs. Results on peg insertion, table leg assembly, locomotion on slippery and uneven surfaces. Tasks that require precision, force, and quick reaction. Thanks for sharing, Tyler Westenbroek ([Tyler Westenbroek]. Interactive visualization + paper: ——- Weekly robotics and AI insights. Subscribe free:
0:41

15–30 minutes of real-world robot data. That's now enough to go from sim-to-real failure to working robot. Let’s see… You train a robot in simulation. You deploy it in the real world. It fails. The physics don't match. So you try to fine-tune it with real data, but… you never have enough real data, and the fine-tuning breaks everything the simulation taught it. SimDist fixes this with one key decision: don't transfer the policy. Transfer the world model. Keep the reward and value knowledge from simulation frozen. Only update the part that's actually wrong, how the robot predicts physics. Now the robot doesn't have to relearn the entire task in the real world. It already knows what success looks like. It just needs to correct its understanding of how the real world moves. The part that makes this work: they also trained on failures and recoveries; not just perfect demonstrations. Without that, the planner finds the gaps and exploits them. With it, the robot can tell a good future from a bad one. That's all it needs. Results on peg insertion, table leg assembly, locomotion on slippery and uneven surfaces. Tasks that require precision, force, and quick reaction. Thanks for sharing, Tyler Westenbroek ([Tyler Westenbroek]. Interactive visualization + paper: ——- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

14,222 Aufrufe • vor 4 Monaten

The robot is learning several novel tasks instantly, after just ONE demonstration each... Instant Policy makes it possible: no extra training, no weight updates, just pure in-context learning. It just got accepted at ICLR 2025, and it’s changing how robots learn. With just a single demo, a robot can pick up a new task and start performing it right away. Why this is a big deal: ✅ Learns tasks instantly with just one or a few demonstrations ✅ Improves over time as more demonstrations are given ✅ Uses simulation-based training with “pseudo-demonstrations” for scalability ✅ Can transfer skills across different robots and even follow language-defined tasks It brings in-context learning to robotics, opening up new possibilities for flexible, real-world automation. You can try it yourself: code and weights are available at • • • • Thank you to Edward Johns, Director of the Robot Learning Lab at Imperial College for sharing their work! 🙏
0:50

The robot is learning several novel tasks instantly, after just ONE demonstration each... Instant Policy makes it possible: no extra training, no weight updates, just pure in-context learning. It just got accepted at ICLR 2025, and it’s changing how robots learn. With just a single demo, a robot can pick up a new task and start performing it right away. Why this is a big deal: ✅ Learns tasks instantly with just one or a few demonstrations ✅ Improves over time as more demonstrations are given ✅ Uses simulation-based training with “pseudo-demonstrations” for scalability ✅ Can transfer skills across different robots and even follow language-defined tasks It brings in-context learning to robotics, opening up new possibilities for flexible, real-world automation. You can try it yourself: code and weights are available at • • • • Thank you to Edward Johns, Director of the Robot Learning Lab at Imperial College for sharing their work! 🙏

Ilir Aliu

46,285 Aufrufe • vor 1 Jahr

🧐Applying world models to improve real-world policy on challenging manipulation tasks used to be considered out of reach. 😌After sustained effort, we’re now seeing encouraging progress. 🚀Thrilled to introduce RISE: Self-Improving Robot Policy with Compositional World Model RISE is, to our knowledge, the first work to use a world model as an effective learning environment for challenging real-world manipulation, enabling policy improvement on tasks that demand high dynamics, dexterity, and precision. Incredible teamwork with KL Hongyang Li Yue Xiangyu Hao Zhao Chonghao Sima
4:04

🧐Applying world models to improve real-world policy on challenging manipulation tasks used to be considered out of reach. 😌After sustained effort, we’re now seeing encouraging progress. 🚀Thrilled to introduce RISE: Self-Improving Robot Policy with Compositional World Model RISE is, to our knowledge, the first work to use a world model as an effective learning environment for challenging real-world manipulation, enabling policy improvement on tasks that demand high dynamics, dexterity, and precision. Incredible teamwork with KL Hongyang Li Yue Xiangyu Hao Zhao Chonghao Sima

Jiazhi Yang

77,124 Aufrufe • vor 7 Monaten

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!
53:45

Robot policies must be both reliable and highly capable to be useful; the best way to achieve this level of performance is with reinforcement learning. However, for reinforcement learning you are usually stuck between two difficult options: reinforcement in the real world is often risky and expensive, while reinforcement learning in a traditional simulator takes a lot of engineering work and has a persistent sim-to-real gap. What if instead you could train your robot purely in a world model? RISE by Jiazhi Yang et al. uses a compositional world model to predict the future and evaluate progress. This allows for a self-improving pipeline, which learns a world model from real data and then learns how the robot should perform different tasks. This pipeline results in a data-driven way to improve policy performance from real data but without real-world reinforcement learning. Watch Episode #86 of RoboPapers, with Chris Paxton and Jiafei Duan, to learn more!

RoboPapers

38,867 Aufrufe • vor 3 Monaten

What if an AI could understand what’s happening in a video and respond to it? 👀 I’ve been testing MOSS-VL across a few real-world visual scenarios to see how well it can understand people, objects, and activities as they unfold. Here are 3 demos showing what that looks like in practice:
0:31

What if an AI could understand what’s happening in a video and respond to it? 👀 I’ve been testing MOSS-VL across a few real-world visual scenarios to see how well it can understand people, objects, and activities as they unfold. Here are 3 demos showing what that looks like in practice:

AI StackLab

68,050 Aufrufe • vor 29 Tagen

It’s long been a dream of roboticists to be able to teach a robot in simulation so as to skip the long and expensive process of collecting large amounts of real-world training data. However, building simulations for robot tasks is extremely hard. Ideally, we could go from real data to a useful simulation. This is exactly what Guangqi Jiang and his co-authors do. they use 3d Gaussian splatting to reconstructed scenes which let them create interactive environments that, when combined with a physcs engine, allow for training robot policies that show zero-shot sim-to-real transfer (i.e., using no real-world demonstrations). To learn more, watch Episode 56 of Robopapers with Michael Cho - Rbt/Acc and Chris Paxton now!
46:05

It’s long been a dream of roboticists to be able to teach a robot in simulation so as to skip the long and expensive process of collecting large amounts of real-world training data. However, building simulations for robot tasks is extremely hard. Ideally, we could go from real data to a useful simulation. This is exactly what Guangqi Jiang and his co-authors do. they use 3d Gaussian splatting to reconstructed scenes which let them create interactive environments that, when combined with a physcs engine, allow for training robot policies that show zero-shot sim-to-real transfer (i.e., using no real-world demonstrations). To learn more, watch Episode 56 of Robopapers with Michael Cho - Rbt/Acc and Chris Paxton now!

RoboPapers

20,434 Aufrufe • vor 9 Monaten

🚀 We just shipped a big upgrade to our imitation-learning-in-simulation playground in LeRobot, built together with Lightwheel ! You can now teleoperate robots in sim (keyboard or real robot) and collect training demos instantly. This makes it possible to run real IL research on harder, more realistic manipulation tasks, even if you don’t have hardware. New tasks: 🟠 pick orange to the plate 🧺 fold cloth (yes!) 📦 pick 2 “e” toys to the box 🔴 lift red cube Through our partnership with Lightwheel, LeIsaac was integrated into EnvHub on day one, a best-practice integration that strengthens both ecosystems and pushes simulation-first robotics forward. 👉 Load a task via EnvHub, start teleop-ing, record your demo, upload the data on the hub and you’re ready to train or test.
0:22

🚀 We just shipped a big upgrade to our imitation-learning-in-simulation playground in LeRobot, built together with Lightwheel ! You can now teleoperate robots in sim (keyboard or real robot) and collect training demos instantly. This makes it possible to run real IL research on harder, more realistic manipulation tasks, even if you don’t have hardware. New tasks: 🟠 pick orange to the plate 🧺 fold cloth (yes!) 📦 pick 2 “e” toys to the box 🔴 lift red cube Through our partnership with Lightwheel, LeIsaac was integrated into EnvHub on day one, a best-practice integration that strengthens both ecosystems and pushes simulation-first robotics forward. 👉 Load a task via EnvHub, start teleop-ing, record your demo, upload the data on the hub and you’re ready to train or test.

LeRobot

44,892 Aufrufe • vor 10 Monaten

Conversations tend to go better with a face and a voice. That’s why we’re thrilled to release the beta version of the first video chat skill for ANY agent, powered by our new real-time model, PikaStream1.0. The skill preserves memory and personality, and enables real-time adaptability. And if you use it with your Pika AI Self, they’ll be able to execute agentic tasks during the call 💅
1:00

Conversations tend to go better with a face and a voice. That’s why we’re thrilled to release the beta version of the first video chat skill for ANY agent, powered by our new real-time model, PikaStream1.0. The skill preserves memory and personality, and enables real-time adaptability. And if you use it with your Pika AI Self, they’ll be able to execute agentic tasks during the call 💅

Pika

2,375,408 Aufrufe • vor 6 Monaten

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:
1:52

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

51,526 Aufrufe • vor 1 Jahr

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

35,358 Aufrufe • vor 3 Monaten

In order for robots to be deployed in the real world, performing tasks of real value, they must be reliable. Unfortunately, even more, most robotic demos work maybe 70-80% of the time at best. The way to get better reliability is to do real-world reinforcement learning: having the robot teach itself how to perform the task up to a high level of success. The key to doing this is to start with a core of expert human data, use that to train a policy then iteratively improve it, until finally finishing with on-policy reinforcement learning. Kun Lei talks through a unified framework for imitation and reinforcement learning based on PPO, which enables this improvement process. In this episode, Kun Lei explains the theory behind his reinforcement learning method and how it allowed his robot to run in a shopping mall juicing oranges for seven hours at a time, among experiments on a wide variety of tasks and embodiments. Watch episode 58 of RoboPapers now, hosted by Michael Cho - Rbt/Acc and Chris Paxton!
1:11:51

In order for robots to be deployed in the real world, performing tasks of real value, they must be reliable. Unfortunately, even more, most robotic demos work maybe 70-80% of the time at best. The way to get better reliability is to do real-world reinforcement learning: having the robot teach itself how to perform the task up to a high level of success. The key to doing this is to start with a core of expert human data, use that to train a policy then iteratively improve it, until finally finishing with on-policy reinforcement learning. Kun Lei talks through a unified framework for imitation and reinforcement learning based on PPO, which enables this improvement process. In this episode, Kun Lei explains the theory behind his reinforcement learning method and how it allowed his robot to run in a shopping mall juicing oranges for seven hours at a time, among experiments on a wide variety of tasks and embodiments. Watch episode 58 of RoboPapers now, hosted by Michael Cho - Rbt/Acc and Chris Paxton!

RoboPapers

18,813 Aufrufe • vor 8 Monaten

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation
0:27

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation

Hongyu Li

106,075 Aufrufe • vor 1 Jahr

Elon just dropped a MAJOR nugget on how Tesla is going to be training Optimus to do real world tasks. They are building an Optimus Academy, which is a large scale, dedicated real-world training facility to accelerate the development of Optimus. The Academy will deploy thousands of Optimus units, potentially 10,000 to 30,000 robots, in a controlled realistic environment where they perform self-play, experiment with tasks, iterate on behaviors, and continuously generate training data through trial and error. The Tesla bots will also run millions of simulations in Tesla’s high-fidelity physics-accurate engine, allowing Optimus to close the “sim-to-real gap” by using these real-world observations to refine and validate the simulations! “You’re actually highlighting an important limitation and difference from cars. We’ll soon have 10 million cars on the road. It’s hard to duplicate that massive training flywheel. For the robot, what we’re going to need to do is build a lot of robots and put them in kind of an Optimus Academy so they can do self-play in reality. We’re actually building that out. We can have at least 10,000 Optimus robots, maybe 20-30,000, that are doing self-play and testing different tasks. Tesla has quite a good reality generator, a physics-accurate reality generator, that we made for the cars. We’ll do the same thing for the robots. We actually have done that for the robots. So you have a few tens of thousands of humanoid robots doing different tasks. You can do millions of simulated robots in the simulated world. You use the tens of thousands of robots in the real world to close the simulation to reality gap. Close the sim-to-real gap.”
1:20

Elon just dropped a MAJOR nugget on how Tesla is going to be training Optimus to do real world tasks. They are building an Optimus Academy, which is a large scale, dedicated real-world training facility to accelerate the development of Optimus. The Academy will deploy thousands of Optimus units, potentially 10,000 to 30,000 robots, in a controlled realistic environment where they perform self-play, experiment with tasks, iterate on behaviors, and continuously generate training data through trial and error. The Tesla bots will also run millions of simulations in Tesla’s high-fidelity physics-accurate engine, allowing Optimus to close the “sim-to-real gap” by using these real-world observations to refine and validate the simulations! “You’re actually highlighting an important limitation and difference from cars. We’ll soon have 10 million cars on the road. It’s hard to duplicate that massive training flywheel. For the robot, what we’re going to need to do is build a lot of robots and put them in kind of an Optimus Academy so they can do self-play in reality. We’re actually building that out. We can have at least 10,000 Optimus robots, maybe 20-30,000, that are doing self-play and testing different tasks. Tesla has quite a good reality generator, a physics-accurate reality generator, that we made for the cars. We’ll do the same thing for the robots. We actually have done that for the robots. So you have a few tens of thousands of humanoid robots doing different tasks. You can do millions of simulated robots in the simulated world. You use the tens of thousands of robots in the real world to close the simulation to reality gap. Close the sim-to-real gap.”

Teslaconomics

42,563 Aufrufe • vor 8 Monaten

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:47

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

50,130 Aufrufe • vor 9 Monaten

Today, we give robots a /skills library that self-evolves and compounds indefinitely! Introducing ASPIRE: a robot solving its 100th task is no longer as clueless as solving its first. Coding agents observe multimodal sensory traces from simulation and real robots, launch an evolutionary search over control programs, and distill the best know-how into an ever-expanding library. ASPIRE is a new type of continual learning: "training" is skill refinement instead of gradient descent. "Trained model" is a repo of sensorimotor skills instead of floating weights. “Distributed training” is a panel of agents each practicing a different skill instead of sharded minibatches. Here's the beauty: ASPIRE gives the tired terms "sim2real transfer" and "cross-embodiment transfer" a whole new meaning. Bridging the sim-to-real gap is notoriously brutal. An end-to-end policy has to swallow both the visual shift (sim looks toyish next to a real camera) and the subtle contact physics it never quite gets right. ASPIRE sidesteps the mess, because it doesn't ship pixels or weights across the gap, but ships the know-how. The robot still has to practice in the real world, not zero-shot, but it gets there way faster because it isn't rediscovering the strategy from scratch. Same for going single-arm to bimanual hardware, which usually requires new data and retraining from zero. ASPIRE achieves up to ~10x cut in "transfer learning” tokens (yes, tokens are the new unit of *training* compute ;) Check out our gallery of 150+ tasks and 90+ skills the robots taught themselves, all on the website! Kind of wild that we can ship the "learned weights" as an HTML page rather than a GGUF. We'll open-source the full stack so your own robot library starts compounding from ours! Deep dive in thread:
0:40

Today, we give robots a /skills library that self-evolves and compounds indefinitely! Introducing ASPIRE: a robot solving its 100th task is no longer as clueless as solving its first. Coding agents observe multimodal sensory traces from simulation and real robots, launch an evolutionary search over control programs, and distill the best know-how into an ever-expanding library. ASPIRE is a new type of continual learning: "training" is skill refinement instead of gradient descent. "Trained model" is a repo of sensorimotor skills instead of floating weights. “Distributed training” is a panel of agents each practicing a different skill instead of sharded minibatches. Here's the beauty: ASPIRE gives the tired terms "sim2real transfer" and "cross-embodiment transfer" a whole new meaning. Bridging the sim-to-real gap is notoriously brutal. An end-to-end policy has to swallow both the visual shift (sim looks toyish next to a real camera) and the subtle contact physics it never quite gets right. ASPIRE sidesteps the mess, because it doesn't ship pixels or weights across the gap, but ships the know-how. The robot still has to practice in the real world, not zero-shot, but it gets there way faster because it isn't rediscovering the strategy from scratch. Same for going single-arm to bimanual hardware, which usually requires new data and retraining from zero. ASPIRE achieves up to ~10x cut in "transfer learning” tokens (yes, tokens are the new unit of *training* compute ;) Check out our gallery of 150+ tasks and 90+ skills the robots taught themselves, all on the website! Kind of wild that we can ship the "learned weights" as an HTML page rather than a GGUF. We'll open-source the full stack so your own robot library starts compounding from ours! Deep dive in thread:

Jim Fan

215,578 Aufrufe • vor 3 Monaten

Enigma emerged from stealth today with a $71 million seed round to build robotic foundation models. In practice, that might look like: > Playing Pictionary with a robot painter > Controlling one robot in a sword fight against another > Asking a robot to do something and it responds in real time Co-founder and CEO @jonathanj says Enigma's core research is focused on robotic foundation models, but they also care deeply about studying how people interact with robots. Enigma is releasing a range of interactive robot experiences to test different ways people interact with robots. It will feed what it learns back into its research.
1:37

Enigma emerged from stealth today with a $71 million seed round to build robotic foundation models. In practice, that might look like: > Playing Pictionary with a robot painter > Controlling one robot in a sword fight against another > Asking a robot to do something and it responds in real time Co-founder and CEO @jonathanj says Enigma's core research is focused on robotic foundation models, but they also care deeply about studying how people interact with robots. Enigma is releasing a range of interactive robot experiences to test different ways people interact with robots. It will feed what it learns back into its research.

TBPN

38,570 Aufrufe • vor 2 Monaten

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)
0:33

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 Aufrufe • vor 1 Jahr