正在加载视频...

视频加载失败

返回首页

What is the best data for training humanoid & robotics foundation models? Pete Florence Pete Florence (CEO Eric Nehrlich, ex-Google DeepMind) dropped his live data tier list in this 7-minute clip on TBPN: - S-tier: Real-world robot experience (especially glove/sensor high-dexterity data) - A/B-tier: Internet/YouTube videos. Surprisingly powerful for...

michellelsun's profile picture

Michelle

7,032 subscribers

14,928 次观看 • 8 天前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

The next bottleneck in AI isn’t model performance. It’s data. From scraped internet content → real-world, consented, high-context data.
2:21

The next bottleneck in AI isn’t model performance. It’s data. From scraped internet content → real-world, consented, high-context data.

Story

10,954 次观看 • 4 天前

The 375ai Mainnet is live! 375ai indexes the real world, transforming video, audio and environmental data into real time intelligence using edge AI. A new data layer for the physical world, now on @Solana
0:55

The 375ai Mainnet is live! 375ai indexes the real world, transforming video, audio and environmental data into real time intelligence using edge AI. A new data layer for the physical world, now on @Solana

375ai

222,148 次观看 • 7 个月前

Leju Robotics has launched a massive 10,000-square-meter data training center in Beijing, aimed at solving the industry’s data shortage and bridging the "reality gap" for AI models. The facility features 1:1 replicas of 16 real-world scenarios across industrial, home, and healthcare sectors. ► Scale: Over 10,000 sq. meters, featuring 16 detailed real-world scenarios. ► Capacity: Annual production capacity of over 6 million high-quality data entries. ► Training: The Kuavo humanoid robot is already training on-site, achieving a 95%+ success rate on practical tasks like sorting, packing, and inspection. ► Quality: Data sourced from real machine runs, certified at a 99% single-data qualification rate.
1:41

Leju Robotics has launched a massive 10,000-square-meter data training center in Beijing, aimed at solving the industry’s data shortage and bridging the "reality gap" for AI models. The facility features 1:1 replicas of 16 real-world scenarios across industrial, home, and healthcare sectors. ► Scale: Over 10,000 sq. meters, featuring 16 detailed real-world scenarios. ► Capacity: Annual production capacity of over 6 million high-quality data entries. ► Training: The Kuavo humanoid robot is already training on-site, achieving a 95%+ success rate on practical tasks like sorting, packing, and inspection. ► Quality: Data sourced from real machine runs, certified at a 99% single-data qualification rate.

RoboHub🤖

43,066 次观看 • 8 个月前

AI is moving beyond the browser and into the real world. The bottleneck? Data. Today we’re announcing a $15M seed round led by a16z crypto to build infra that collects, curates, and licenses high-quality data for physical AI. Incubated by and built on Story.
0:37

AI is moving beyond the browser and into the real world. The bottleneck? Data. Today we’re announcing a $15M seed round led by a16z crypto to build infra that collects, curates, and licenses high-quality data for physical AI. Incubated by and built on Story.

Poseidon

291,571 次观看 • 11 个月前

🤖 INTERESTING: RLWRLD is collecting real-world dexterity data using MANUS gloves to train physical AI systems on human hand and finger movements.
0:21

🤖 INTERESTING: RLWRLD is collecting real-world dexterity data using MANUS gloves to train physical AI systems on human hand and finger movements.

Cointelegraph

26,524 次观看 • 1 个月前

Origami Robotics is building high-DOF robotic hands with in-joint motors and a co-designed data-collection glove to eliminate the embodiment gap by collecting high-quality, real-world data at scale. Congrats on the launch, Quanting Xie and Ryan Xie!
1:09

Origami Robotics is building high-DOF robotic hands with in-joint motors and a co-designed data-collection glove to eliminate the embodiment gap by collecting high-quality, real-world data at scale. Congrats on the launch, Quanting Xie and Ryan Xie!

Y Combinator

88,257 次观看 • 3 个月前

1/ The data parsing market for robotics is just emerging, and it’s already valued in the tens of billions. Data parsing to build data models and software for Robotics is a $20B market in 2024 and is expected to reach $150B+ by 2034 Cherry AI Robotics has built a data parsing tool for robotics, which goes live next week, and will effectively make this project one of the pioneers in data parsing for robotics.
0:25

1/ The data parsing market for robotics is just emerging, and it’s already valued in the tens of billions. Data parsing to build data models and software for Robotics is a $20B market in 2024 and is expected to reach $150B+ by 2034 Cherry AI Robotics has built a data parsing tool for robotics, which goes live next week, and will effectively make this project one of the pioneers in data parsing for robotics.

CherryAI Robotics

90,786 次观看 • 9 个月前

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 次观看 • 1 年前

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 🔗
0:29

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

13,408 次观看 • 4 个月前

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)
1:15

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

44,227 次观看 • 1 年前

So I heard we need more data for robot learning :) Purely real world teleop is expensive and slow, making large scale data collection challenging. I’ve been excited about getting more data into robot learning, going beyond just real-world teleop data. To this end, we’ve been scaling up data generation with RL in realistic simulations generated on the fly from crowdsourced videos. Enables realistic data collection, much more cheaply than purely real world teleop. Importantly, data collection becomes even*cheaper* with more environments, allowing training with over 100x more data. Transfers to real robots for generalizable manipulation. A 🧵 (1/N)
0:41

So I heard we need more data for robot learning :) Purely real world teleop is expensive and slow, making large scale data collection challenging. I’ve been excited about getting more data into robot learning, going beyond just real-world teleop data. To this end, we’ve been scaling up data generation with RL in realistic simulations generated on the fly from crowdsourced videos. Enables realistic data collection, much more cheaply than purely real world teleop. Importantly, data collection becomes even*cheaper* with more environments, allowing training with over 100x more data. Transfers to real robots for generalizable manipulation. A 🧵 (1/N)

Abhishek Gupta

13,336 次观看 • 1 年前

.Sureform (YC X25) connects real-world workplaces with robotics labs to collect task-specific training data for robot foundation models. Congrats on the launch, Ananth Kashyap!
0:34

.Sureform (YC X25) connects real-world workplaces with robotics labs to collect task-specific training data for robot foundation models. Congrats on the launch, Ananth Kashyap!

Y Combinator

35,876 次观看 • 3 个月前

Watch ROVR🌐｜Data-Driven Planet’s pitch from the Crypto’s Got Talent Semifinals: ROVR is building a global network of LiDAR-powered devices to collect high-fidelity 3D data for AI and autonomous systems. Deployed across Asia and beyond, this is real-world DePIN for real-world AI. ↓
17:40

Watch ROVR🌐｜Data-Driven Planet’s pitch from the Crypto’s Got Talent Semifinals: ROVR is building a global network of LiDAR-powered devices to collect high-fidelity 3D data for AI and autonomous systems. Deployed across Asia and beyond, this is real-world DePIN for real-world AI. ↓

IoTeX

27,582 次观看 • 1 年前

the micro1 robotics lab: real world data for intelligent models that co-exist in the physical world. we’re in-the-wild across 75 countries in 6,000+ unique environments collecting data. diverse movements, objects, and settings. the future of AI is as human as you can imagine. join us to start training robots today (link in comments).
1:05

the micro1 robotics lab: real world data for intelligent models that co-exist in the physical world. we’re in-the-wild across 75 countries in 6,000+ unique environments collecting data. diverse movements, objects, and settings. the future of AI is as human as you can imagine. join us to start training robots today (link in comments).

Ali Ansari

207,061 次观看 • 3 个月前

The "sim-to-real" gap just got a massive 2,300m² bridge. 🤖🏗️ NEURA Robotics and TUM are launching the TUM RoboGym at Munich Airport—Europe’s largest training center for Physical AI. Starting mid-2026, a fleet of humanoid robots will begin real-world training to fuel the Neuraverse ecosystem with high-quality, sovereign data.
1:41

The "sim-to-real" gap just got a massive 2,300m² bridge. 🤖🏗️ NEURA Robotics and TUM are launching the TUM RoboGym at Munich Airport—Europe’s largest training center for Physical AI. Starting mid-2026, a fleet of humanoid robots will begin real-world training to fuel the Neuraverse ecosystem with high-quality, sovereign data.

Humanoids daily

25,560 次观看 • 3 个月前

Robots and AI cannot hear the world - Yet. They are starving for fuel, and the internet’s data is already scraped dry. AI needs real-world audio data - at massive scale. Silencio is expanding its vision: from the largest noise data bank to The World’s Ears for AI and Robotics. 📅On September 17th: The world gets ears. Reserve your exclusive keynote preview:
0:25

Robots and AI cannot hear the world - Yet. They are starving for fuel, and the internet’s data is already scraped dry. AI needs real-world audio data - at massive scale. Silencio is expanding its vision: from the largest noise data bank to The World’s Ears for AI and Robotics. 📅On September 17th: The world gets ears. Reserve your exclusive keynote preview:

Silencio | Voice Data for AI

95,420 次观看 • 9 个月前

Unreal Engine 5 is becoming the go-to platform for robotics teams, not just as a simulator but as a synthetic data factory powering the next generation of robotics systems🦾 From synthetic data to sim-to-real transfer, read our piece on how UE5 is quietly reshaping how robots learn to see, move, and act in the real world:
0:36

Unreal Engine 5 is becoming the go-to platform for robotics teams, not just as a simulator but as a synthetic data factory powering the next generation of robotics systems🦾 From synthetic data to sim-to-real transfer, read our piece on how UE5 is quietly reshaping how robots learn to see, move, and act in the real world:

Unreal Engine

61,187 次观看 • 2 个月前

Physical Intelligence co-founder, Sergey Levine: The best synthetic experience now comes from strong models, not hand-coded physics Models can capture finer details than humans, but still need real-world experience "information in, capability out" Whether through simulation or direct data, what matters is what the model can do
0:56

Physical Intelligence co-founder, Sergey Levine: The best synthetic experience now comes from strong models, not hand-coded physics Models can capture finer details than humans, but still need real-world experience "information in, capability out" Whether through simulation or direct data, what matters is what the model can do

Haider.

14,197 次观看 • 9 个月前

AI only exists today because it is trained on real data. The better the data input, The better the AI output. We capture incredibly high-quality raw data of real humans, in 3D, from all angles. You do the Math ✍🏻 Our data can vastly improve and expand the capabilities of AI
1:04

AI only exists today because it is trained on real data. The better the data input, The better the AI output. We capture incredibly high-quality raw data of real humans, in 3D, from all angles. You do the Math ✍🏻 Our data can vastly improve and expand the capabilities of AI

Metahero.io

24,527 次观看 • 2 年前

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.
0:58

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.

The Humanoid Hub

46,715 次观看 • 1 年前