正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

The next breakthrough for robot VLMs isn’t a bigger model 🧠 it’s memory. RAL accepted ✅ We show that a robot can ground a VLM using its own real-world experience, not extra training. EXPTEACH/PRAGMABOT = self-generated memory + closed-loop reflection: Fail → diagnose → replan → succeed → store... show more

Haitham Bou Ammar

5,423 subscribers

12,206 次观看 • 6 个月前 •via X (Twitter)

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

Hey #NeuraxonMini is literally out! , we manage to "transplant" a Neuraxon 2 bioinspired #AI brain to a physical robot the #SpheroMini moving from our last Scientific Paper (link bellow) by David Vivancos - e/acc & Jose Sánchez for Qubic #OpenScience hybridized with #Aigarth to the real World. First you need a Sphero Education Mini robot about 50$ Then you can try the first cool demos at Hugging Face: 1.- Neuraxon2MiniControl to drive the sphero robot 2.- Neuraxon2MiniWrite to write letters or words with physical moves of the sphero robot using Neuraxon Video Tutorials on youtube later today. Why this matters? Remember we are not building "dead" LLMs we are building #AliveAIs and for that we need to explore how it behaves in reality, from how it learns to how it fails, and what better way that in the emerging field of #robotics , time will tell if your next #HumanoidRobot have a #Neuraxon brain... Read the Paper: Explore the Neuraxon code here: Are you ready for #TrueAI ?

Hey #NeuraxonMini is literally out! , we manage to "transplant" a Neuraxon 2 bioinspired #AI brain to a physical robot the #SpheroMini moving from our last Scientific Paper (link bellow) by David Vivancos - e/acc & Jose Sánchez for Qubic #OpenScience hybridized with #Aigarth to the real World. First you need a Sphero Education Mini robot about 50$ Then you can try the first cool demos at Hugging Face: 1.- Neuraxon2MiniControl to drive the sphero robot 2.- Neuraxon2MiniWrite to write letters or words with physical moves of the sphero robot using Neuraxon Video Tutorials on youtube later today. Why this matters? Remember we are not building "dead" LLMs we are building #AliveAIs and for that we need to explore how it behaves in reality, from how it learns to how it fails, and what better way that in the emerging field of #robotics , time will tell if your next #HumanoidRobot have a #Neuraxon brain... Read the Paper: Explore the Neuraxon code here: Are you ready for #TrueAI ?

David Vivancos - e/acc

29,293 次观看 • 6 个月前

Imagine you go to a store and you want to buy candy. The shopkeeper knows you're a real kid because they can see you standing right there. Now imagine you send a robot to buy candy for you. The shopkeeper looks at the robot and thinks: wait, who sent this? Is this robot allowed to buy candy? What if someone else's robot pretends to be yours and steals your candy money? That's basically what's happening with AI right now. Companies like Visa let people buy things all over the world. But now, smart computer robots (AI agents) want to buy things too. Shop around, compare prices, even pay for stuff. Visa looked at this and said: nope, not yet. Because they have no way to check if the robot is real, who it belongs to, or if it's allowed to spend that money. The problem is that all the rules we have for checking identity - showing your ID, scanning your face, typing your password - only work for humans. Robots can't do any of that. Worse, bad robots can actually copy and fake human identities really well. So Evin McMullen evin, Billions Network co-founder and CEO, says we need a new kind of ID system. One where you can prove something is true without showing all your private stuff. Like proving you're tall enough for a ride without telling anyone your exact height. That's called zero-knowledge proof. And for the robots specifically, we need something called KYA - Know Your Agent. It's like giving every robot its own ID card that says: this is who I am, this is what I'm allowed to do, and this is the human responsible for me. Until we build that, the robot economy can't really get going. Here is Evin’s Thought Leader article at Silicon Valleys Journal

Imagine you go to a store and you want to buy candy. The shopkeeper knows you're a real kid because they can see you standing right there. Now imagine you send a robot to buy candy for you. The shopkeeper looks at the robot and thinks: wait, who sent this? Is this robot allowed to buy candy? What if someone else's robot pretends to be yours and steals your candy money? That's basically what's happening with AI right now. Companies like Visa let people buy things all over the world. But now, smart computer robots (AI agents) want to buy things too. Shop around, compare prices, even pay for stuff. Visa looked at this and said: nope, not yet. Because they have no way to check if the robot is real, who it belongs to, or if it's allowed to spend that money. The problem is that all the rules we have for checking identity - showing your ID, scanning your face, typing your password - only work for humans. Robots can't do any of that. Worse, bad robots can actually copy and fake human identities really well. So Evin McMullen evin, Billions Network co-founder and CEO, says we need a new kind of ID system. One where you can prove something is true without showing all your private stuff. Like proving you're tall enough for a ride without telling anyone your exact height. That's called zero-knowledge proof. And for the robots specifically, we need something called KYA - Know Your Agent. It's like giving every robot its own ID card that says: this is who I am, this is what I'm allowed to do, and this is the human responsible for me. Until we build that, the robot economy can't really get going. Here is Evin’s Thought Leader article at Silicon Valleys Journal

Billions Network

21,775 次观看 • 6 个月前

🚨China vs USA robots Chinese robot: $24,000 American robot: $39,100 Pricing policy isn't everything. If we look at real capabilities, the balance of power looks completely different 🇨🇳 The robot demonstrates incredible agility, confidently performs a backflip, moves dynamically, and dances Maximum flexibility, high tech at an affordable price, and readiness for real-world tasks 🇺🇸 The robot, with a higher price tag, currently looks more modest, struggling even with basic rectilinear walking An obvious overpayment for brand and marketing, while Chinese engineers are outpacing competitors in software and hardware The main paradox of the current moment is that Chinese companies focus on rapidly cheapening hardware and mass production, turning robotics into an accessible mass-market consumer product. Meanwhile, the US traditionally tries to enter the market through fundamental neural networks and a universal AI brain However, in the coming years, we will see the inevitable convergence of these approaches: the winner in this race will not be the one who simply teaches a robot to do a backflip or walk straight, but the one who first combines Chinese cheap and adaptive mechatronics with advanced American language and multimodal models And what do you think, who will ultimately set the global standard in robotics East or West?

Bober_smart

34,127 次观看 • 11 天前

CES was an enormous success. We saw thousands of people and a few dogs interacting with our agents and robots. We booked this booth at the last minute, mainly to book next year’s booth. Truly extraordinary what our team was able to put together on such short notice. Next year we reserved one of the largest booths at CES in the highest traffic location for AI and Robotics. We plan to unveil something extraordinary. Until the next show, we will be debuting the first custom models that will emerge from the first humanoid robotics hot rod shop. These models will be walking around, talking and interacting at tradeshows all over the world over the next few months. We are in the process of acquiring every humanoid robot available on the market to modify these as well. Stay tuned for more updates, the future is here and it’s rapidly accelerating.

CES was an enormous success. We saw thousands of people and a few dogs interacting with our agents and robots. We booked this booth at the last minute, mainly to book next year’s booth. Truly extraordinary what our team was able to put together on such short notice. Next year we reserved one of the largest booths at CES in the highest traffic location for AI and Robotics. We plan to unveil something extraordinary. Until the next show, we will be debuting the first custom models that will emerge from the first humanoid robotics hot rod shop. These models will be walking around, talking and interacting at tradeshows all over the world over the next few months. We are in the process of acquiring every humanoid robot available on the market to modify these as well. Stay tuned for more updates, the future is here and it’s rapidly accelerating.

Parallel

16,874 次观看 • 1 年前

THIS AIBO ISN’T A TOY. IT’S THE MOST ADVANCED ROBOT PET EVER BUILT. 🐕🤖 Sony didn’t just build a gadget — they built something that runs on 22-axis lifelike movement, recognizes your face, and develops its own “personality” over time based on how you interact with it. Here’s what makes it wild: 🐾 Learns and adapts — no two Aibos behave exactly the same after a few months of use 🐾 Cloud-connected personality — its behavior profile lives on Sony’s servers, evolving with every interaction 🐾 Free developer API with visual programming — which has made it a favorite tool in robotics classrooms worldwide 🐾 Real-world use cases far beyond “cute robot dog”: home companionship, senior care and emotional support, STEM education, even security patrol duties No vet bills. No allergies. No mess. Just a machine that’s been engineered to feel less like a product and more like a living companion. We used to think “robot pets” meant clunky toys with blinking lights. Sony quietly spent years proving the whole category could feel real. The future of companionship isn’t fully human anymore. And it’s already sitting on someone’s couch right now.

DN_DEGEN

114,038 次观看 • 23 天前

Force feedback demo Force feedback is when joystick is pushing on your hand when something is pushing on the robot arm. Feeling the force - so much helpful to control the robot, that done well it allows you to do tasks even without visual feed. You can make an experiment: close your eyes - you can easily get the headphones out of the case. Also, visual information is often not enough. For example, you're trying to pull out a usb connector, but you pull it at the wrong angle, causing it to get stuck. Visually, nothing changes, but the pressure is intense and you can break the connector. Surgical robots have been using force feedback for years, and there are also 3D styluses which use this feature, proving that the technology works and is useful. But in modern robots with AI, it's hardly ever implemented. Although it's useful for both teleoperation and AI model. That's one of the reasons why we are building our robotic arms starting with off the shelf motors rather than taking the whole off the shelf arm. There are still a range of easy wins that can be made iterating robot hardware.

Force feedback demo Force feedback is when joystick is pushing on your hand when something is pushing on the robot arm. Feeling the force - so much helpful to control the robot, that done well it allows you to do tasks even without visual feed. You can make an experiment: close your eyes - you can easily get the headphones out of the case. Also, visual information is often not enough. For example, you're trying to pull out a usb connector, but you pull it at the wrong angle, causing it to get stuck. Visually, nothing changes, but the pressure is intense and you can break the connector. Surgical robots have been using force feedback for years, and there are also 3D styluses which use this feature, proving that the technology works and is useful. But in modern robots with AI, it's hardly ever implemented. Although it's useful for both teleoperation and AI model. That's one of the reasons why we are building our robotic arms starting with off the shelf motors rather than taking the whole off the shelf arm. There are still a range of easy wins that can be made iterating robot hardware.

Igor Kulakov

18,773 次观看 • 1 年前

𝗥𝗼𝗯𝗼𝘁𝘀 𝗱𝗼𝗻’𝘁 𝗻𝗲𝗲𝗱 𝗺𝗼𝗿𝗲 𝗱𝗲𝗺𝗼𝗻𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻𝘀. 𝗧𝗵𝗲𝘆 𝗻𝗲𝗲𝗱 𝘁𝗼 𝗹𝗲𝗮𝗿𝗻 𝗳𝗿𝗼𝗺 𝗳𝗮𝗶𝗹𝘂𝗿𝗲 — 𝗮𝗳𝘁𝗲𝗿 𝘄𝗮𝘁𝗰𝗵𝗶𝗻𝗴 𝗵𝘂𝗺𝗮𝗻𝘀. Most robot learning systems assume failure is the end of learning. In our new work, we study whether robots can improve after deployment by learning from their own failures, without any human intervention, teleoperation, or corrective labels. The key idea is simple: human videos contain structure about how the world works. We use them to learn cross-embodiment representations of action, dynamics, and value, enabling a shared predictive space between human behavior and robot experience. This allows a new learning loop: 👉 pretrain on human videos 👉 deploy robot policy 👉 observe failures 👉 reinterpret failures using human priors 👉 improve autonomously We evaluate this across 7 real-world manipulation tasks, showing: 📈 40% → 81% success rate 🏆 Strong improvements over π0.6 RECAP and RISE ✔️ Zero human intervention during post-deployment improvement 🧬 Generalizes across robot embodiments and policy backbones A key finding is that explicit failure repair significantly outperforms failure reweighting, yielding substantially larger gains under identical data conditions (+25 pts vs +5 pts on the same π0.5 base policy). Overall, the results suggest a shift in how we think about robot learning: Human videos are not only for pretraining policies. They can provide the structure needed for continual self-improvement after deployment. 📄 Paper: 🌐 Project: I am grateful for working with the fantastic leads Hanzhi Chen and Anran Zhang, and our collaborators Simon Schaefer, Kejia Chen, Shi Chen, Daniel Cremers. Special thanks to Stefan Leutenegger for co-advising this project with me. ETH Zürich TU München Microsoft Check out Hanzhi's 🧵 for more details

𝗥𝗼𝗯𝗼𝘁𝘀 𝗱𝗼𝗻’𝘁 𝗻𝗲𝗲𝗱 𝗺𝗼𝗿𝗲 𝗱𝗲𝗺𝗼𝗻𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻𝘀. 𝗧𝗵𝗲𝘆 𝗻𝗲𝗲𝗱 𝘁𝗼 𝗹𝗲𝗮𝗿𝗻 𝗳𝗿𝗼𝗺 𝗳𝗮𝗶𝗹𝘂𝗿𝗲 — 𝗮𝗳𝘁𝗲𝗿 𝘄𝗮𝘁𝗰𝗵𝗶𝗻𝗴 𝗵𝘂𝗺𝗮𝗻𝘀. Most robot learning systems assume failure is the end of learning. In our new work, we study whether robots can improve after deployment by learning from their own failures, without any human intervention, teleoperation, or corrective labels. The key idea is simple: human videos contain structure about how the world works. We use them to learn cross-embodiment representations of action, dynamics, and value, enabling a shared predictive space between human behavior and robot experience. This allows a new learning loop: 👉 pretrain on human videos 👉 deploy robot policy 👉 observe failures 👉 reinterpret failures using human priors 👉 improve autonomously We evaluate this across 7 real-world manipulation tasks, showing: 📈 40% → 81% success rate 🏆 Strong improvements over π0.6 RECAP and RISE ✔️ Zero human intervention during post-deployment improvement 🧬 Generalizes across robot embodiments and policy backbones A key finding is that explicit failure repair significantly outperforms failure reweighting, yielding substantially larger gains under identical data conditions (+25 pts vs +5 pts on the same π0.5 base policy). Overall, the results suggest a shift in how we think about robot learning: Human videos are not only for pretraining policies. They can provide the structure needed for continual self-improvement after deployment. 📄 Paper: 🌐 Project: I am grateful for working with the fantastic leads Hanzhi Chen and Anran Zhang, and our collaborators Simon Schaefer, Kejia Chen, Shi Chen, Daniel Cremers. Special thanks to Stefan Leutenegger for co-advising this project with me. ETH Zürich TU München Microsoft Check out Hanzhi's 🧵 for more details

Oier Mees

12,379 次观看 • 2 个月前

This work makes a humanoid robot do simple parkour moves by looking with a depth camera and choosing the right move on the fly. The big deal is that it turns lots of small human moves into long, real-time robot behavior, without hand-coding every transition or retraining for each new course. A humanoid robot is usually good at steady walking, but it often fails when it has to do fast moves like jumping up, vaulting, or rolling, and then keep going to the next obstacle. The hard part is that you cannot easily collect training data for every possible obstacle shape, distance, and mistake, so robots end up learning a few moves that only work in a narrow setup. This work starts from short clips of real human parkour moves, like stepping over, vaulting, climbing, and rolling. It uses motion matching, which is basically a smart “pick the next clip that fits best right now” search, to stitch those short clips into a long, smooth plan that looks like a human doing a whole course. Then it trains a controller with reinforcement learning (RL), which means the robot learns by trial and error to copy that plan while staying balanced and not falling. After training separate expert controllers for different moves, it compresses them into 1 controller that uses only onboard depth sensing and a simple “go this fast in this direction” command. In real tests on a Unitree G1 humanoid, it can clear multiple obstacles in a row, adapt when obstacles get moved, and climb a wall up to 1.25m.

This work makes a humanoid robot do simple parkour moves by looking with a depth camera and choosing the right move on the fly. The big deal is that it turns lots of small human moves into long, real-time robot behavior, without hand-coding every transition or retraining for each new course. A humanoid robot is usually good at steady walking, but it often fails when it has to do fast moves like jumping up, vaulting, or rolling, and then keep going to the next obstacle. The hard part is that you cannot easily collect training data for every possible obstacle shape, distance, and mistake, so robots end up learning a few moves that only work in a narrow setup. This work starts from short clips of real human parkour moves, like stepping over, vaulting, climbing, and rolling. It uses motion matching, which is basically a smart “pick the next clip that fits best right now” search, to stitch those short clips into a long, smooth plan that looks like a human doing a whole course. Then it trains a controller with reinforcement learning (RL), which means the robot learns by trial and error to copy that plan while staying balanced and not falling. After training separate expert controllers for different moves, it compresses them into 1 controller that uses only onboard depth sensing and a simple “go this fast in this direction” command. In real tests on a Unitree G1 humanoid, it can clear multiple obstacles in a row, adapt when obstacles get moved, and climb a wall up to 1.25m.

Rohan Paul

37,121 次观看 • 6 个月前

If you’ve been ignoring Axis Robotics because it looks like another random points farm, read this. In the last 48 hours, Unitree said robotics is approaching its "ChatGPT moment", while the chairman of ACE Robotics believes it could happen by the end of 2027. But for robots to reach that level, they need a crazy amount of training data. That’s exactly what Axis is building. You control simulated robots directly from your browser, complete simple tasks and earn points. Those movements also help create data that can be used to train real robots. And this isn’t some tiny experiment anymore: - $12M raised - 123K+ contributors - 3M+ robot trajectories - Community data already used to train a real robot So yeah, we’re basically farming a potential airdrop while teaching our future robot servants how to work 😂 If you haven’t started yet, it’s completely free. You only need a tiny amount of gas on Base to sign your completed tasks. ✅ Start farming Axis points: Important: Sign every completed task from the History page, otherwise you won’t receive the points.

If you’ve been ignoring Axis Robotics because it looks like another random points farm, read this. In the last 48 hours, Unitree said robotics is approaching its "ChatGPT moment", while the chairman of ACE Robotics believes it could happen by the end of 2027. But for robots to reach that level, they need a crazy amount of training data. That’s exactly what Axis is building. You control simulated robots directly from your browser, complete simple tasks and earn points. Those movements also help create data that can be used to train real robots. And this isn’t some tiny experiment anymore: - $12M raised - 123K+ contributors - 3M+ robot trajectories - Community data already used to train a real robot So yeah, we’re basically farming a potential airdrop while teaching our future robot servants how to work 😂 If you haven’t started yet, it’s completely free. You only need a tiny amount of gas on Base to sign your completed tasks. ✅ Start farming Axis points: Important: Sign every completed task from the History page, otherwise you won’t receive the points.

Pranjal Bora 🧭

29,321 次观看 • 5 天前

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and *self-divergence.* In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and self-divergence. In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.

Minqi Jiang

41,099 次观看 • 11 个月前

Furniture assembly is the task everyone name-drops and nobody actually attempts at real scale. Every demo I have seen is a scaled down IKEA leg or a single arm on a toy chair. This paper does it properly, real scale, bimanual, up to 7 subtasks and 1,550 control steps per episode, and it is validated on a real Kinova Gen3, not just in sim. That real-robot number is the one that matters: only a 16 percent drop on the hardest task going from simulation to hardware. That is a small enough gap to take seriously, and it did not happen by accident. They built a VR teleoperation rig specifically for coordinated dual-arm collection, because generic single-arm teleop setups do not capture the coordination real assembly needs, and the model predicts a continuous progress signal alongside the action chunk rather than a discrete subtask label, letting it auto-transition and catch drift before it compounds into total failure. The simulation ablation is what got them there, 48 to 80 percent over baselines, with another 21 points from their perception and control design study alone, but that is groundwork, not the headline. Watch the video, there is a clip of the robot misgrasping the seat panel, reopening the gripper, and regrasping on its own. That is not scripted recovery behaviour, it emerged from training, and it emerged on hardware. Excellent work from the team from Mitsubishi Electric Research Laboratories, with Oxford and UNC Chapel Hill Clinical Laboratory Science. Video and project page in comments. #Robotics #Manipulation #VLA

Furniture assembly is the task everyone name-drops and nobody actually attempts at real scale. Every demo I have seen is a scaled down IKEA leg or a single arm on a toy chair. This paper does it properly, real scale, bimanual, up to 7 subtasks and 1,550 control steps per episode, and it is validated on a real Kinova Gen3, not just in sim. That real-robot number is the one that matters: only a 16 percent drop on the hardest task going from simulation to hardware. That is a small enough gap to take seriously, and it did not happen by accident. They built a VR teleoperation rig specifically for coordinated dual-arm collection, because generic single-arm teleop setups do not capture the coordination real assembly needs, and the model predicts a continuous progress signal alongside the action chunk rather than a discrete subtask label, letting it auto-transition and catch drift before it compounds into total failure. The simulation ablation is what got them there, 48 to 80 percent over baselines, with another 21 points from their perception and control design study alone, but that is groundwork, not the headline. Watch the video, there is a clip of the robot misgrasping the seat panel, reopening the gripper, and regrasping on its own. That is not scripted recovery behaviour, it emerged from training, and it emerged on hardware. Excellent work from the team from Mitsubishi Electric Research Laboratories, with Oxford and UNC Chapel Hill Clinical Laboratory Science. Video and project page in comments. #Robotics #Manipulation #VLA

Stephen James

14,952 次观看 • 1 个月前

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

Jonathan Stephens

17,712 次观看 • 7 个月前

Robotics keeps hitting the same wall. Single task RL works, but... it does not scale to hundreds of tasks or new embodiments. This new paper looks like a real step toward fixing that. The team introduces MMBench, a benchmark with 200 tasks across many domains and robots, and Newt, a language conditioned world model trained online across all 200 tasks at once. The simple idea behind Newt: The model learns from demos to get the right priors It trains across many tasks through online interaction It uses language to ground the goal It adapts fast when a new task shows up What stood out to me: ✅ One model trained on 200 tasks at the same time ✅ Language conditioned control for both states and RGB ✅ Better data efficiency than strong baselines ✅ Strong open loop control ✅ Fast adaptation to new tasks and embodiments ✅ Full release of 200 checkpoints, 4000 demos, code, and benchmark This is a good push toward general control instead of one model per task. If you want the full paper: Project page: —- Weekly robotics and AI insights. Subscribe free:

Robotics keeps hitting the same wall. Single task RL works, but... it does not scale to hundreds of tasks or new embodiments. This new paper looks like a real step toward fixing that. The team introduces MMBench, a benchmark with 200 tasks across many domains and robots, and Newt, a language conditioned world model trained online across all 200 tasks at once. The simple idea behind Newt: The model learns from demos to get the right priors It trains across many tasks through online interaction It uses language to ground the goal It adapts fast when a new task shows up What stood out to me: ✅ One model trained on 200 tasks at the same time ✅ Language conditioned control for both states and RGB ✅ Better data efficiency than strong baselines ✅ Strong open loop control ✅ Fast adaptation to new tasks and embodiments ✅ Full release of 200 checkpoints, 4000 demos, code, and benchmark This is a good push toward general control instead of one model per task. If you want the full paper: Project page: —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

70,090 次观看 • 9 个月前

🚨 BREAKING: NVIDIA just announced the Isaac GR00T Reference Humanoid Robot. The first fully open humanoid robot reference design built on Jetson Thor, and it's going straight to the world's top research institutions. This is Jensen Huang's bet on open physical AI infrastructure. The hardware stack is serious: → Unitree H2 Plus chassis, 6 feet tall, 150 pounds, 31 degrees of freedom → Sharpa Wave tactile five-finger hands, 22 degrees of freedom, bringing total to 75 across the full body → NVIDIA Jetson AGX Thor onboard compute, 2,070 FP4 teraflops of AI performance, 128GB unified memory → Multi-view sensing, stereo head camera, wrist cameras, IMU Alongside this announcement, Unitree also introduced the H2 Plus as a standalone product, a frontier humanoid combining Unitree's own body, Sharpa's five-finger hands and NVIDIA Robotics Jetson Thor compute into one fully integrated research platform. The full Isaac GR00T software stack ships with it, teleoperation for data capture, open foundation models, Isaac Sim for training, Isaac Lab for evaluation, and accelerated ROS middleware for deployment. The complete loop from data to real-world robot in one unified platform. ETH Zürich, Stanford Robotics Center, UC San Diego and Ai2 are already on board as launch research partners. NVIDIA Robotics did to AI what it's now doing to robotics, build the platform, open the ecosystem, let the world build on top of it. Whoever owns the infrastructure layer wins. NVIDIA knows this better than anyone. 👀 Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

🚨 BREAKING: NVIDIA just announced the Isaac GR00T Reference Humanoid Robot. The first fully open humanoid robot reference design built on Jetson Thor, and it's going straight to the world's top research institutions. This is Jensen Huang's bet on open physical AI infrastructure. The hardware stack is serious: → Unitree H2 Plus chassis, 6 feet tall, 150 pounds, 31 degrees of freedom → Sharpa Wave tactile five-finger hands, 22 degrees of freedom, bringing total to 75 across the full body → NVIDIA Jetson AGX Thor onboard compute, 2,070 FP4 teraflops of AI performance, 128GB unified memory → Multi-view sensing, stereo head camera, wrist cameras, IMU Alongside this announcement, Unitree also introduced the H2 Plus as a standalone product, a frontier humanoid combining Unitree's own body, Sharpa's five-finger hands and NVIDIA Robotics Jetson Thor compute into one fully integrated research platform. The full Isaac GR00T software stack ships with it, teleoperation for data capture, open foundation models, Isaac Sim for training, Isaac Lab for evaluation, and accelerated ROS middleware for deployment. The complete loop from data to real-world robot in one unified platform. ETH Zürich, Stanford Robotics Center, UC San Diego and Ai2 are already on board as launch research partners. NVIDIA Robotics did to AI what it's now doing to robotics, build the platform, open the ecosystem, let the world build on top of it. Whoever owns the infrastructure layer wins. NVIDIA knows this better than anyone. 👀 Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

16,062 次观看 • 2 个月前

It's 2030 and you are reviewing humanoid robots. A Tesla. A Google. An Apple. An OpenAI. A Meta. A Figure. And a bunch of Chinese-made ones. Which one is best, and why? I think the Tesla understands the world much better. Why? There were eight Teslas around me on the freeway today. Start there. No other robot company has that data. But my robot is parked at the local high school twice a day. Its cameras see humans in all of our weirdness. How we move. Where we go. Where we walk. Who we talk with. What you are wearing. Whether your hair was combed this morning. That data will lead to robotics breakthroughs. Apple might keep up with its Vision Pro data, but it is too freaked out by the privacy implications of using said data. (On the front are six cameras and a couple of TOF -- Time Of Flight -- sensors that can see everything in your home in great detail). Google has a lot of data, for sure. All my: 1. Email. 2. Calendars. 3. Photos. 4. TV watching behavior. 5. Contacts. 6. Documents and spreadsheets. 7. Files. 8. Location data. So I expect Google's robot will be attractive to many. But how do you see the others shake out over the next five years? Make some guesses. But remember what an AI pioneer told me years ago about AI: it's all about the data. The Chinese ones have huge advantages: the Chinese have more data on their citizens, and many more citizens to boot AND they can make robots cheaper than we can. But now that you know OpenAI is building its own robot you have caught wind of what I've heard from many in San Francisco and Silicon Valley: that humanoid robots are the real prize of AI and will be highly profitable for those that can make them and find customers willing to buy them. Here, too, I learned long ago never to bet against Elon Musk. Will you?

It's 2030 and you are reviewing humanoid robots. A Tesla. A Google. An Apple. An OpenAI. A Meta. A Figure. And a bunch of Chinese-made ones. Which one is best, and why? I think the Tesla understands the world much better. Why? There were eight Teslas around me on the freeway today. Start there. No other robot company has that data. But my robot is parked at the local high school twice a day. Its cameras see humans in all of our weirdness. How we move. Where we go. Where we walk. Who we talk with. What you are wearing. Whether your hair was combed this morning. That data will lead to robotics breakthroughs. Apple might keep up with its Vision Pro data, but it is too freaked out by the privacy implications of using said data. (On the front are six cameras and a couple of TOF -- Time Of Flight -- sensors that can see everything in your home in great detail). Google has a lot of data, for sure. All my: 1. Email. 2. Calendars. 3. Photos. 4. TV watching behavior. 5. Contacts. 6. Documents and spreadsheets. 7. Files. 8. Location data. So I expect Google's robot will be attractive to many. But how do you see the others shake out over the next five years? Make some guesses. But remember what an AI pioneer told me years ago about AI: it's all about the data. The Chinese ones have huge advantages: the Chinese have more data on their citizens, and many more citizens to boot AND they can make robots cheaper than we can. But now that you know OpenAI is building its own robot you have caught wind of what I've heard from many in San Francisco and Silicon Valley: that humanoid robots are the real prize of AI and will be highly profitable for those that can make them and find customers willing to buy them. Here, too, I learned long ago never to bet against Elon Musk. Will you?

Robert Scoble

33,804 次观看 • 1 年前

AI Is Moving Beyond “Generating Videos” — Toward “Generating Worlds” Over the past two years, AI video models have advanced at an astonishing pace. From Runway and Pika to Sora and Veo, AI-generated videos have become increasingly realistic and more consistent with the physical laws of the real world. Many people believe the next objective is simply to generate videos that are longer, sharper, and more lifelike. But if we take a step back, we can see that the real transformation is not happening in video itself. It is happening in world models. What Is a World Model? In 1943, psychologist Kenneth Craik proposed an idea that would influence artificial intelligence research for decades. He argued that the human brain does not merely react to the outside world. Instead, it maintains an internal model of how the world works. Because we have this internal model, we can predict the outcome of an action before we actually take it. Before crossing a road, we estimate whether a car will pass by. Before catching a ball, we predict its trajectory. These abilities come from continuously simulating the world in our minds, rather than relying entirely on trial and error. This idea later became known by a more formal term: World Model. A world model does not describe a single image or a fixed video clip. It is an internal representation capable of continuously simulating the rules and dynamics of the real world. Why Is AI Research Turning Toward World Models? Because predicting “what comes next” is becoming increasingly central to how AI systems work. Language models predict the next token. Image models predict the next step in the denoising process. Video models predict the next frame. A world model, however, attempts to predict something broader: What should the world look like in the next moment? In 2018, David Ha and Jürgen Schmidhuber proposed in their paper World Models that an intelligent agent could first learn a model of the world, and then use that internal model to plan its actions. The Dreamer series later demonstrated that many complex tasks could be learned by training agents inside an “imagined world.” At the same time, the development of video models such as Sora and Veo led researchers to another realization: A model capable of continuously generating video has already learned, at least implicitly, many of the rules governing the real world. As a result, these two research directions have gradually begun to converge. But Video Is Not Yet a World This is where the distinction is often misunderstood. For a world model to support meaningful real-time interaction, it must solve several critical problems. Most video models today are essentially answering one question: What should the next frame look like? A true world model needs to answer much more: What happens if I take one step forward? If I walk behind a building and then return, will the building still be there? If I suddenly change the camera angle, will the entire space remain consistent? If I enter a command such as: “Summon a dragon.” Will the world respond immediately? In other words, a world model must do more than generate content. It must understand space. It must understand time. It must understand causality. And it must understand interaction. Moving from watching to participating is where the real difficulty of world models begins. World Models Are Entering the Interactive Era One of the latest attempts in this direction is Alaya World, recently open-sourced by Alaya World, or Alaya Lab. Instead of generating a fixed video clip, it generates a world that users can explore in real time. Users can begin with text, an image, or a video, enter the generated scene, move freely through it, and introduce new prompts at any moment during generation. The world responds immediately. According to the publicly released information, Alaya World provides: Real-time streaming generation at 720p and 24 FPS Stable continuous exploration for more than one minute The ability to switch prompts and trigger skills or events during generation Model weights and inference code released under the Apache 2.0 License Training code and datasets planned for future release What makes these capabilities important is not simply the technical specifications. It is that the generated “world” can now support continuous interaction. The official demo shows that users can genuinely control, transform, and explore the generated environment. AI Is Evolving From a Tool Into an Environment Over the past few years, most discussions around AI have focused on content generation. Generating text. Generating images. Generating videos. But world models raise a fundamentally different question: Can AI generate an environment that people can inhabit, explore, and continuously evolve? If the answer is yes, the impact will extend far beyond video generation. Game development, robotics training, embodied intelligence, digital twins, virtual production, and many other fields could be transformed by the development of world models. World models are still at a very early stage. Yet from Craik’s proposal of an internal mental model more than eighty years ago to the emergence of today’s interactive world-generation systems, a clear evolutionary path is beginning to take shape. Perhaps what AI is ultimately learning has never been limited to images, videos, or language. Perhaps it is learning the world itself. References GitHub: Technical Report:

AI Is Moving Beyond “Generating Videos” — Toward “Generating Worlds” Over the past two years, AI video models have advanced at an astonishing pace. From Runway and Pika to Sora and Veo, AI-generated videos have become increasingly realistic and more consistent with the physical laws of the real world. Many people believe the next objective is simply to generate videos that are longer, sharper, and more lifelike. But if we take a step back, we can see that the real transformation is not happening in video itself. It is happening in world models. What Is a World Model? In 1943, psychologist Kenneth Craik proposed an idea that would influence artificial intelligence research for decades. He argued that the human brain does not merely react to the outside world. Instead, it maintains an internal model of how the world works. Because we have this internal model, we can predict the outcome of an action before we actually take it. Before crossing a road, we estimate whether a car will pass by. Before catching a ball, we predict its trajectory. These abilities come from continuously simulating the world in our minds, rather than relying entirely on trial and error. This idea later became known by a more formal term: World Model. A world model does not describe a single image or a fixed video clip. It is an internal representation capable of continuously simulating the rules and dynamics of the real world. Why Is AI Research Turning Toward World Models? Because predicting “what comes next” is becoming increasingly central to how AI systems work. Language models predict the next token. Image models predict the next step in the denoising process. Video models predict the next frame. A world model, however, attempts to predict something broader: What should the world look like in the next moment? In 2018, David Ha and Jürgen Schmidhuber proposed in their paper World Models that an intelligent agent could first learn a model of the world, and then use that internal model to plan its actions. The Dreamer series later demonstrated that many complex tasks could be learned by training agents inside an “imagined world.” At the same time, the development of video models such as Sora and Veo led researchers to another realization: A model capable of continuously generating video has already learned, at least implicitly, many of the rules governing the real world. As a result, these two research directions have gradually begun to converge. But Video Is Not Yet a World This is where the distinction is often misunderstood. For a world model to support meaningful real-time interaction, it must solve several critical problems. Most video models today are essentially answering one question: What should the next frame look like? A true world model needs to answer much more: What happens if I take one step forward? If I walk behind a building and then return, will the building still be there? If I suddenly change the camera angle, will the entire space remain consistent? If I enter a command such as: “Summon a dragon.” Will the world respond immediately? In other words, a world model must do more than generate content. It must understand space. It must understand time. It must understand causality. And it must understand interaction. Moving from watching to participating is where the real difficulty of world models begins. World Models Are Entering the Interactive Era One of the latest attempts in this direction is Alaya World, recently open-sourced by Alaya World, or Alaya Lab. Instead of generating a fixed video clip, it generates a world that users can explore in real time. Users can begin with text, an image, or a video, enter the generated scene, move freely through it, and introduce new prompts at any moment during generation. The world responds immediately. According to the publicly released information, Alaya World provides: Real-time streaming generation at 720p and 24 FPS Stable continuous exploration for more than one minute The ability to switch prompts and trigger skills or events during generation Model weights and inference code released under the Apache 2.0 License Training code and datasets planned for future release What makes these capabilities important is not simply the technical specifications. It is that the generated “world” can now support continuous interaction. The official demo shows that users can genuinely control, transform, and explore the generated environment. AI Is Evolving From a Tool Into an Environment Over the past few years, most discussions around AI have focused on content generation. Generating text. Generating images. Generating videos. But world models raise a fundamentally different question: Can AI generate an environment that people can inhabit, explore, and continuously evolve? If the answer is yes, the impact will extend far beyond video generation. Game development, robotics training, embodied intelligence, digital twins, virtual production, and many other fields could be transformed by the development of world models. World models are still at a very early stage. Yet from Craik’s proposal of an internal mental model more than eighty years ago to the emergence of today’s interactive world-generation systems, a clear evolutionary path is beginning to take shape. Perhaps what AI is ultimately learning has never been limited to images, videos, or language. Perhaps it is learning the world itself. References GitHub: Technical Report:

雪踏乌云

113,347 次观看 • 1 个月前

Jeff Bezos just told you exactly how to price AI. Nobody listened. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Three words that reprice the entire technology sector. The iPhone was a vertical. One product. One new market. Electricity was a horizontal. One substrate that rewired every market on Earth. Wall Street is pricing AI like it is the next iPhone. Bezos is telling you it is the next electrical grid. Right now, thousands of companies are trying to sell AI as a product. A feature. A tool. A subscription tier. Every single one of them will be priced to zero. You do not sell a horizontal layer. You do not compete with it. You build on top of it or you disappear beneath it. For a century, entire industries survived on one thing. Complexity. The friction of navigating law, medicine, logistics, finance. That was the moat. If you could not memorize the maze, you could not compete. A horizontal layer does not navigate the maze. It dissolves the walls. Electricity did not compete with the candle industry. It erased the need for one. The most dangerous part of a horizontal shift is how quiet it is. It moves underneath the economy. The surface looks normal. Revenue still holds. Every day you operate on the old substrate, you accumulate a debt you cannot see and cannot repay. The internet repriced distribution. AI is repricing cognition itself. When intelligence becomes a utility that runs through the walls of every company on Earth, the premium on human expertise does not erode. It evaporates. This is not a disruption. Disruptions replace products. This replaces the ground you are standing on.

Jeff Bezos just told you exactly how to price AI. Nobody listened. Bezos: “AI is real and it is going to change every industry. In fact it’s a very unusual technology in that regard in that it’s a horizontal enabling layer.” Horizontal enabling layer. Three words that reprice the entire technology sector. The iPhone was a vertical. One product. One new market. Electricity was a horizontal. One substrate that rewired every market on Earth. Wall Street is pricing AI like it is the next iPhone. Bezos is telling you it is the next electrical grid. Right now, thousands of companies are trying to sell AI as a product. A feature. A tool. A subscription tier. Every single one of them will be priced to zero. You do not sell a horizontal layer. You do not compete with it. You build on top of it or you disappear beneath it. For a century, entire industries survived on one thing. Complexity. The friction of navigating law, medicine, logistics, finance. That was the moat. If you could not memorize the maze, you could not compete. A horizontal layer does not navigate the maze. It dissolves the walls. Electricity did not compete with the candle industry. It erased the need for one. The most dangerous part of a horizontal shift is how quiet it is. It moves underneath the economy. The surface looks normal. Revenue still holds. Every day you operate on the old substrate, you accumulate a debt you cannot see and cannot repay. The internet repriced distribution. AI is repricing cognition itself. When intelligence becomes a utility that runs through the walls of every company on Earth, the premium on human expertise does not erode. It evaporates. This is not a disruption. Disruptions replace products. This replaces the ground you are standing on.

Dustin

541,957 次观看 • 4 个月前

The term "continual learning" has become overloaded if you see it as an ML problem. One classic thread is about memorization: regularization-based continual learning methods, such as EWC, MAS, and SI, estimate which parameters mattered for previous tasks and resist changing them too much. One modern thread is about adaptation: test-time training and inference-time learning methods, such as TTT, adapt part of the model on the incoming test stream before making predictions. These are sometimes discussed as separate threads. But in modern scalable architectures, I think they are better seen as complementary constraints: a model that learns quickly at test time also benefits from a mechanism for deciding what not to forget. In our #ECCV2026 paper, we study this in large-scale 4D reconstruction: how to build fast spatial memory that can adapt over long observation streams while reducing collapse and forgetting. Instead of using fully plastic test-time updates, we stabilize fast-weight adaptation with an elastic prior that balances adaptation and memory. Key ideas: - Elastic Test-Time Training: Fisher-weighted consolidation for fast-weight updates - EMA anchor weights that provide a moving reference for stability - Chunk-by-chunk inference for long 3D/4D observation streams We show that this scales across large 3D/4D pretraining settings, including both LRM-style and LVSM-style models, and improves reconstruction across benchmarks including Stereo4D, NVIDIA, and DL3DV-140. We release model checkpoints across different design choices: resolution, post-training curriculum, and whether the model uses an explicit 4DGS intermediate representation. - Homepage: - Paper: - Code: - Models: This work is co-led with Xueyang Yu, contributed by Haoyu Zhen Yuncong Yang, and advised by Michigan SLED Lab Chuang Gan.

The term "continual learning" has become overloaded if you see it as an ML problem. One classic thread is about memorization: regularization-based continual learning methods, such as EWC, MAS, and SI, estimate which parameters mattered for previous tasks and resist changing them too much. One modern thread is about adaptation: test-time training and inference-time learning methods, such as TTT, adapt part of the model on the incoming test stream before making predictions. These are sometimes discussed as separate threads. But in modern scalable architectures, I think they are better seen as complementary constraints: a model that learns quickly at test time also benefits from a mechanism for deciding what not to forget. In our #ECCV2026 paper, we study this in large-scale 4D reconstruction: how to build fast spatial memory that can adapt over long observation streams while reducing collapse and forgetting. Instead of using fully plastic test-time updates, we stabilize fast-weight adaptation with an elastic prior that balances adaptation and memory. Key ideas: - Elastic Test-Time Training: Fisher-weighted consolidation for fast-weight updates - EMA anchor weights that provide a moving reference for stability - Chunk-by-chunk inference for long 3D/4D observation streams We show that this scales across large 3D/4D pretraining settings, including both LRM-style and LVSM-style models, and improves reconstruction across benchmarks including Stereo4D, NVIDIA, and DL3DV-140. We release model checkpoints across different design choices: resolution, post-training curriculum, and whether the model uses an explicit 4DGS intermediate representation. - Homepage: - Paper: - Code: - Models: This work is co-led with Xueyang Yu, contributed by Haoyu Zhen Yuncong Yang, and advised by Michigan SLED Lab Chuang Gan.

Martin Ziqiao Ma

33,767 次观看 • 2 个月前