Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Introducing RoboCat, a new AI model designed to operate multiple robots. 🤖 It learns to solve new tasks on different robotic arms with as few as 100 demonstrations - and improves skills from self-generated training data. Find out more:

Google DeepMind

1,458,263 subscribers

410,272 Aufrufe • vor 3 Jahren •via X (Twitter)

Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

10 Kommentare

Profilbild von Google DeepMind

Google DeepMindvor 3 Jahren

🟥 Actions like shape matching and building structures are complex for robots. They have to master how to control each joint, coordinate movements, and know the physics of their surroundings. This embodied intelligence comes naturally to humans but it’s very difficult for AI.

Profilbild von Google DeepMind

Google DeepMindvor 3 Jahren

🔵 RoboCat learns faster than other state-of-the-art models because it draws from a large, diverse dataset. This feature will advance robotics research as it reduces the need for human-supervised training - a key step to creating general-purpose robots:

Profilbild von Bobcat

Bobcatvor 3 Jahren

@DeepMind But what of the cat operating all this

Profilbild von Kye Gomez (ultra/acc)

Kye Gomez (ultra/acc)vor 3 Jahren

@DeepMind IS IT OPEN SOURCE?

Profilbild von co-varun

co-varunvor 3 Jahren

@DeepMind Whaaaaaaaat! Nooo wayyyyy!

Profilbild von Ben Schulz

Ben Schulzvor 3 Jahren

@DeepMind This will be huge for the eventual autonomous construction vehicles.

Profilbild von Viral Vora- Towards Super intelligence

Viral Vora- Towards Super intelligencevor 3 Jahren

@DeepMind A tweet that should have been a presentation

Profilbild von Aaron Francis

Aaron Francisvor 3 Jahren

@DeepMind @KennethCassel sure you've seen this, but tagging you anyway! Seems like you're in a good space 🔥

Profilbild von Astur81

Astur81vor 3 Jahren

@DeepMind @DotCSV

Profilbild von heyheymymy

heyheymymyvor 3 Jahren

@DeepMind For our generation, when we get older, we may be taken care by robots. Every household will have a robot, just like every household will have a car.

Ähnliche Videos

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:

Google DeepMind

537,765 Aufrufe • vor 2 Jahren

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! 🙏

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

Training robot foundation models faces two key hurdles: how to get enough data to train an effective model, and how to make sure that new skills can be acquired quickly. The team at Rhoda AI believes that the answer is training Direct Video Action models from web data. Web data is plentiful, to the point where Rhoda can train their base model on hundreds of years of video data. And then, with the addition of robot data, they can quickly adapt it to new tasks with as little as 20 hours of in-domain data, performing complex, multi-step manipulation tasks with their purpose-built video foundation model. Tongzhou Mu 🤖🦾🦿 Eric Chan and Changan Chen joined us to talk more about their approach. Watch Episode #79 of RoboPapers, with Michael Cho - Rbt/Acc, Chris Paxton, and Jiafei Duan, to learn more!

Training robot foundation models faces two key hurdles: how to get enough data to train an effective model, and how to make sure that new skills can be acquired quickly. The team at Rhoda AI believes that the answer is training Direct Video Action models from web data. Web data is plentiful, to the point where Rhoda can train their base model on hundreds of years of video data. And then, with the addition of robot data, they can quickly adapt it to new tasks with as little as 20 hours of in-domain data, performing complex, multi-step manipulation tasks with their purpose-built video foundation model. Tongzhou Mu 🤖🦾🦿 Eric Chan and Changan Chen joined us to talk more about their approach. Watch Episode #79 of RoboPapers, with Michael Cho - Rbt/Acc, Chris Paxton, and Jiafei Duan, to learn more!

RoboPapers

24,352 Aufrufe • vor 1 Monat

Data is 👑 Our foundation model learns from globally diverse, unlabelled data. Without telling it how to drive or naming different scene elements, our AI learns patterns and behaviours. Our results this week demonstrate how rapidly our AI can learn new behaviours, finding that with hundreds of hours worth of country-specific data, our foundation model learns new skills, like driving on the right in the US! See how our Wayve AI Driver improves driving in the US with data from 0-10-100-250-500 hours worth of additional US-specific data. 👇 By leveraging diverse data, we are building a highly generalizable AI that enables safe, adaptable, and scalable autonomy—we’ve taken it to the US and Germany but stay tuned for where is next 🌎⁉️

Data is 👑 Our foundation model learns from globally diverse, unlabelled data. Without telling it how to drive or naming different scene elements, our AI learns patterns and behaviours. Our results this week demonstrate how rapidly our AI can learn new behaviours, finding that with hundreds of hours worth of country-specific data, our foundation model learns new skills, like driving on the right in the US! See how our Wayve AI Driver improves driving in the US with data from 0-10-100-250-500 hours worth of additional US-specific data. 👇 By leveraging diverse data, we are building a highly generalizable AI that enables safe, adaptable, and scalable autonomy—we’ve taken it to the US and Germany but stay tuned for where is next 🌎⁉️

Alex Kendall

17,178 Aufrufe • vor 1 Jahr

Scientists developed HumanX, a system that lets humanoid robots learn new skills by watching just one short human video. It learns directly from demonstrations instead of large datasets or complex training rewards. Using this, the Unitree G1 humanoid robot learned 10 different real-world skills without extra training.

Scientists developed HumanX, a system that lets humanoid robots learn new skills by watching just one short human video. It learns directly from demonstrations instead of large datasets or complex training rewards. Using this, the Unitree G1 humanoid robot learned 10 different real-world skills without extra training.

Space and Technology

27,686 Aufrufe • vor 2 Monaten

🤖 How can robot policies zero-shot generalize to any new environment and any new object? Introducing our new project: 🚀Data Scaling Laws in Imitation Learning for Robotic Manipulation🚀—bringing us closer to the dream of having robots work as waiters in hot pot restaurants! 🍲

🤖 How can robot policies zero-shot generalize to any new environment and any new object? Introducing our new project: 🚀Data Scaling Laws in Imitation Learning for Robotic Manipulation🚀—bringing us closer to the dream of having robots work as waiters in hot pot restaurants! 🍲

Yang Gao

124,995 Aufrufe • vor 1 Jahr

We’re bringing powerful AI directly onto robots with Gemini Robotics On-Device. 🤖 It’s our first vision-language-action model to help make robots faster, highly efficient, and adaptable to new tasks and environments - without needing a constant internet connection. 🧵

We’re bringing powerful AI directly onto robots with Gemini Robotics On-Device. 🤖 It’s our first vision-language-action model to help make robots faster, highly efficient, and adaptable to new tasks and environments - without needing a constant internet connection. 🧵

Google DeepMind

819,249 Aufrufe • vor 1 Jahr

Manus AI released Scheduled Tasks 2.0, introducing web apps that can self-update their data on a schedule, task continuation, and new ways to observe ongoing tasks. > Scheduled Tasks is now available to all Manus users.

Manus AI released Scheduled Tasks 2.0, introducing web apps that can self-update their data on a schedule, task continuation, and new ways to observe ongoing tasks. > Scheduled Tasks is now available to all Manus users.

🚨 AI News | TestingCatalog

21,024 Aufrufe • vor 1 Monat

Excited to share our latest progress on DYNA-1 pre-training! 🤖 The base model now can perform diverse, dexterous tasks (laundry folding, package sorting, …) without any post-training, even in unseen environments. This powerful base also allows extremely efficient fine-tuning to ~100% success on challenging new tasks with as little as 1 hour of data! 🤯 Watch it master two of them: cup stacking & celery chopping on repeat, no failures. 👇

Excited to share our latest progress on DYNA-1 pre-training! 🤖 The base model now can perform diverse, dexterous tasks (laundry folding, package sorting, …) without any post-training, even in unseen environments. This powerful base also allows extremely efficient fine-tuning to ~100% success on challenging new tasks with as little as 1 hour of data! 🤯 Watch it master two of them: cup stacking & celery chopping on repeat, no failures. 👇

Dyna Robotics

66,619 Aufrufe • vor 7 Monaten

Can we build a generalist robotic policy that doesn’t just memorize training data and regurgitate it during test time, but instead remembers past actions as memory and conditions its decisions on them?🤖💡 Introducing SAM2Act—a multi-view robotic transformer-based policy that integrates a visual foundation model with a memory architecture for robotic manipulation. Project page: 🧵👇

Can we build a generalist robotic policy that doesn’t just memorize training data and regurgitate it during test time, but instead remembers past actions as memory and conditions its decisions on them?🤖💡 Introducing SAM2Act—a multi-view robotic transformer-based policy that integrates a visual foundation model with a memory architecture for robotic manipulation. Project page: 🧵👇

Jiafei Duan

87,573 Aufrufe • vor 1 Jahr

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

100,274 Aufrufe • vor 3 Monaten

The $10k robot arms that can learn from a handful of demos: Most robots still lose the cost equation. They’re slow, expensive, and take months to deploy, which means… they only make sense for huge factories. The The Robot Learning Company (YC X25) is building affordable, general-purpose robot arms that can pick up new tasks from just a few human demonstrations. Why it matters: ✅ Under $10k per pair of stationary arms ✅ Adapts to new tasks with imitation + reinforcement learning ✅ Deployable in days, not months From small manufacturers to service providers, this makes robotic labor accessible to companies that could never justify it before. Saying hi to founder Jannik Grothusen 👋 More here:

The $10k robot arms that can learn from a handful of demos: Most robots still lose the cost equation. They’re slow, expensive, and take months to deploy, which means… they only make sense for huge factories. The The Robot Learning Company (YC X25) is building affordable, general-purpose robot arms that can pick up new tasks from just a few human demonstrations. Why it matters: ✅ Under $10k per pair of stationary arms ✅ Adapts to new tasks with imitation + reinforcement learning ✅ Deployable in days, not months From small manufacturers to service providers, this makes robotic labor accessible to companies that could never justify it before. Saying hi to founder Jannik Grothusen 👋 More here:

Ilir Aliu - eu/acc

36,336 Aufrufe • vor 10 Monaten

Can GPTs generate infinite and diverse data for robotics? Introducing RoboGen, a generative robotic agent that keeps proposing new tasks, creating corresponding environments and acquiring novel skills autonomously! code: 👇🧵 (better with audio)

Can GPTs generate infinite and diverse data for robotics? Introducing RoboGen, a generative robotic agent that keeps proposing new tasks, creating corresponding environments and acquiring novel skills autonomously! code: 👇🧵 (better with audio)

Zhou Xian

124,192 Aufrufe • vor 2 Jahren

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

51,490 Aufrufe • vor 1 Jahr

Big News! Meta just released Segment Anything, a new AI model that can "cut out" any object, in any image/video, with a single click. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks.

Big News! Meta just released Segment Anything, a new AI model that can "cut out" any object, in any image/video, with a single click. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks.

Lior Alexander

290,190 Aufrufe • vor 3 Jahren

Behold. Poop detection from Matic Robots with their new robotic vacuum Much more on a new episode of Core Memory that you can find down below

Behold. Poop detection from Matic Robots with their new robotic vacuum Much more on a new episode of Core Memory that you can find down below

Ashlee Vance

13,387 Aufrufe • vor 10 Monaten

AI is only as intelligent as the data it learns from. And data is only as good as the humans behind it. Introducing Humyn Labs. Where human intelligence is not just a step, it’s the filter. 👇

AI is only as intelligent as the data it learns from. And data is only as good as the humans behind it. Introducing Humyn Labs. Where human intelligence is not just a step, it’s the filter. 👇

Humyn Labs

17,166 Aufrufe • vor 3 Monaten

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:

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

22,812 Aufrufe • vor 7 Monaten

As AI agents get better at computer and tool use, or writing code on the fly for a task, we're going to be able to solve much broader domains of knowledge work. Here's an example of Box AI with the new Claude Skills to generate a clean powerpoint file from existing data.

As AI agents get better at computer and tool use, or writing code on the fly for a task, we're going to be able to solve much broader domains of knowledge work. Here's an example of Box AI with the new Claude Skills to generate a clean powerpoint file from existing data.

Aaron Levie

30,235 Aufrufe • vor 8 Monaten

Introducing Rabbithole: A new visual way to dive into topics, powered by AI-generated follow-up questions that help you learn deeply. Rabbithole is designed to expand your curiosity. As someone whose curiosity has only expanded with AI, I’ve often wanted a more visually expansive way to learn about anything on my mind - without having to switch back and forth between tabs or chats. Try it out (on desktop) -

Introducing Rabbithole: A new visual way to dive into topics, powered by AI-generated follow-up questions that help you learn deeply. Rabbithole is designed to expand your curiosity. As someone whose curiosity has only expanded with AI, I’ve often wanted a more visually expansive way to learn about anything on my mind - without having to switch back and forth between tabs or chats. Try it out (on desktop) -

Mohammed Ibrahim

318,148 Aufrufe • vor 1 Jahr