Loading video...

Video Failed to Load

Go Home

For 41 years, to find shortest paths in a graph, the Dijkstra’s algorithm, was seen as the best possible way. Not any longer: Now a team from Tsinghua University has beaten it. They created the first faster algorithm for directed shortest paths since 1984. • Faster than Dijkstra on...

IlirAliu_'s profile picture

Ilir Aliu

53,922 subscribers

95,270 views • 6 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Almost every robot you see... runs on this equation. Not AI. Not machine learning. PID. For over 100 years, this simple control algorithm has been quietly keeping robots balanced, drones stable, industrial machines precise, and even rockets on course. Most people never hear about it. Yet without PID, many of today's robots wouldn't even stand upright. If you could only learn ONE control algorithm in robotics... this would be it. What's the most impressive application of PID you've seen? 🎥 Media: medcorreia ( Instagram ) ⚠️ This content is shared for informational purposes only. CTO Robotics Media is a media platform and does not own or develop the technology shown. Credit belongs to the original creators.
0:18

Almost every robot you see... runs on this equation. Not AI. Not machine learning. PID. For over 100 years, this simple control algorithm has been quietly keeping robots balanced, drones stable, industrial machines precise, and even rockets on course. Most people never hear about it. Yet without PID, many of today's robots wouldn't even stand upright. If you could only learn ONE control algorithm in robotics... this would be it. What's the most impressive application of PID you've seen? 🎥 Media: medcorreia ( Instagram ) ⚠️ This content is shared for informational purposes only. CTO Robotics Media is a media platform and does not own or develop the technology shown. Credit belongs to the original creators.

CTO ROBOTICS Media

45,912 views • 1 month ago

Its not every day you wake up to find that the Pope has made your lifes work the central focus of his papacy: “Disarming AI means freeing it from the mentality of “armed” competition [..] This entails a race for ever more powerful algorithms and larger datasets, driven by the desire to secure geopolitical or commercial dominance." - POPE LEO XIV, May 2026 Here, “disarmed” means “neutralised” in the sense that this should not be a differentiator The Innovation Game (TIG) was created to keep data and algorithms open, in order to prevent monopolistic control It's not just an aspiration, it's an economic mechanism that makes open data and open algorithms the rational economic choice • All algorithms are published openly by TIG • If you are willing to make the data you process with an algorithm open, you can use it free-of-charge • Alternatively, if you would like to keep this data private, there is a fee to pay for using the algorithm • All fees are used to fund more open innovation Its an elegant, global, self-reinforcing engine The logical end point of monopoly is that innovation stops We cannot allow that to happen Pope Leo XIV I would be grateful for your thoughts on The Innovation Game
0:18

Its not every day you wake up to find that the Pope has made your lifes work the central focus of his papacy: “Disarming AI means freeing it from the mentality of “armed” competition [..] This entails a race for ever more powerful algorithms and larger datasets, driven by the desire to secure geopolitical or commercial dominance." - POPE LEO XIV, May 2026 Here, “disarmed” means “neutralised” in the sense that this should not be a differentiator The Innovation Game (TIG) was created to keep data and algorithms open, in order to prevent monopolistic control It's not just an aspiration, it's an economic mechanism that makes open data and open algorithms the rational economic choice • All algorithms are published openly by TIG • If you are willing to make the data you process with an algorithm open, you can use it free-of-charge • Alternatively, if you would like to keep this data private, there is a fee to pay for using the algorithm • All fees are used to fund more open innovation Its an elegant, global, self-reinforcing engine The logical end point of monopoly is that innovation stops We cannot allow that to happen Pope Leo XIV I would be grateful for your thoughts on The Innovation Game

John Fletcher (𝔦, 𝔦)

11,722 views • 2 months ago

AI in robotics gets all the attention right now, but sometimes the most interesting work is very practical. Viet built a small vision system that counts potatoes on a conveyor belt. No giant dataset. No huge model. Just a clear problem and a smart setup. He used Ultralytics’ ObjectCounter, trained a tiny YOLO11 nano model, and because there was no potato dataset, he annotated a single frame with SAM 2 and trained from that. One frame. Still works across the whole video. It is a good reminder that useful AI in industry often looks like this. Focused. Lightweight. Solves a real task. If you work in manufacturing or robotics, these small systems are usually the fastest wins. They save time, reduce errors, and do not need massive infrastructure. Nice work, Viet. His projects: —- Weekly robotics and AI insights. Subscribe free:
0:13

AI in robotics gets all the attention right now, but sometimes the most interesting work is very practical. Viet built a small vision system that counts potatoes on a conveyor belt. No giant dataset. No huge model. Just a clear problem and a smart setup. He used Ultralytics’ ObjectCounter, trained a tiny YOLO11 nano model, and because there was no potato dataset, he annotated a single frame with SAM 2 and trained from that. One frame. Still works across the whole video. It is a good reminder that useful AI in industry often looks like this. Focused. Lightweight. Solves a real task. If you work in manufacturing or robotics, these small systems are usually the fastest wins. They save time, reduce errors, and do not need massive infrastructure. Nice work, Viet. His projects: —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

1,675,497 views • 8 months ago

AI in robotics gets all the attention right now, but sometimes the most interesting work is very practical. Viet built a small vision system that counts potatoes on a conveyor belt. No giant dataset. No huge model. Just a clear problem and a smart setup. He used Ultralytics’ ObjectCounter, trained a tiny YOLO11 nano model, and because there was no potato dataset, he annotated a single frame with SAM 2 and trained from that. One frame. Still works across the whole video. It is a good reminder that useful AI in industry often looks like this. Focused. Lightweight. Solves a real task. If you work in manufacturing or robotics, these small systems are usually the fastest wins. They save time, reduce errors, and do not need massive infrastructure. Nice work, Viet. His projects: ——- Weekly robotics and AI insights. Subscribe free:
0:13

AI in robotics gets all the attention right now, but sometimes the most interesting work is very practical. Viet built a small vision system that counts potatoes on a conveyor belt. No giant dataset. No huge model. Just a clear problem and a smart setup. He used Ultralytics’ ObjectCounter, trained a tiny YOLO11 nano model, and because there was no potato dataset, he annotated a single frame with SAM 2 and trained from that. One frame. Still works across the whole video. It is a good reminder that useful AI in industry often looks like this. Focused. Lightweight. Solves a real task. If you work in manufacturing or robotics, these small systems are usually the fastest wins. They save time, reduce errors, and do not need massive infrastructure. Nice work, Viet. His projects: ——- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

196,825 views • 1 day ago

Multi-axis 3D printing with curved layers! 🖨️ Researchers from the The University of Manchester introduced a neural network-based computational pipeline as a representation-agnostic slicer for multi-axis 3D printing. Traditional 3D printing works like stacking pancakes, flat layers on top of each other. 🥞 This often requires temporary support structures that get thrown away after printing, wastes material, and creates weaker parts. Multi-axis 3D printing can print along curved paths that follow the object's natural shape. This eliminates support structures and makes stronger parts. But figuring out these curved paths is mathematically complex, you need to avoid collisions, respect what the printer can physically do, and optimize for strength. The neural network solves this automatically. It learns to create a "field" around the object, then extracts curved printing paths from this field. Because the entire process is differentiable (translation for non-math specialists, meaning you can optimize it end-to-end), the AI can directly optimize for manufacturing goals like "no support structures needed" and "make it as strong as possible." Here's the project: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:14

Multi-axis 3D printing with curved layers! 🖨️ Researchers from the The University of Manchester introduced a neural network-based computational pipeline as a representation-agnostic slicer for multi-axis 3D printing. Traditional 3D printing works like stacking pancakes, flat layers on top of each other. 🥞 This often requires temporary support structures that get thrown away after printing, wastes material, and creates weaker parts. Multi-axis 3D printing can print along curved paths that follow the object's natural shape. This eliminates support structures and makes stronger parts. But figuring out these curved paths is mathematically complex, you need to avoid collisions, respect what the printer can physically do, and optimize for strength. The neural network solves this automatically. It learns to create a "field" around the object, then extracts curved printing paths from this field. Because the entire process is differentiable (translation for non-math specialists, meaning you can optimize it end-to-end), the AI can directly optimize for manufacturing goals like "no support structures needed" and "make it as strong as possible." Here's the project: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

57,048 views • 4 months ago

Most imitation learning policies break when the camera moves or the robot changes. NOT THIS ONE 👇 [📍 Bookmark for later ] A new 3D scene representation encoder, tackles this by enabling zero-shot generalization to unseen embodiments and viewpoints… And it works with any IL algorithm. The trick? •Use a 2D foundation model to extract semantic features •Lift them into 3D space for localization (not semantics) •Condition the IL policy on this spatially grounded vector Across 93 simulated and 6 real tasks, Adapt3R: ✅ Maintains IL performance on LIBERO & MimicGen benchmarks ✅ Outperforms DP3 and 3D Diffuser Actor in most settings ✅ Holds >80% success on LIBERO even with large camera rotations Thanks for sharing this, Animesh Garg & Albert Wilcox! 📍Paper: Website: Code:
0:14

Most imitation learning policies break when the camera moves or the robot changes. NOT THIS ONE 👇 [📍 Bookmark for later ] A new 3D scene representation encoder, tackles this by enabling zero-shot generalization to unseen embodiments and viewpoints… And it works with any IL algorithm. The trick? •Use a 2D foundation model to extract semantic features •Lift them into 3D space for localization (not semantics) •Condition the IL policy on this spatially grounded vector Across 93 simulated and 6 real tasks, Adapt3R: ✅ Maintains IL performance on LIBERO & MimicGen benchmarks ✅ Outperforms DP3 and 3D Diffuser Actor in most settings ✅ Holds >80% success on LIBERO even with large camera rotations Thanks for sharing this, Animesh Garg & Albert Wilcox! 📍Paper: Website: Code:

Ilir Aliu

12,178 views • 11 months ago

Most humanoid projects talk about real work. Very few last an hour on a real line. This week I saw a case that matters for anyone building robots, perception, or physical AI. Kinisi deployed its first mobile manipulation system into a live recycling facility. Not a demo. Not a staged test. A real production line with real output pressure. Why this matters if you want robotics to deliver real value on your floor: • Handles mixed glass with random poses and no fixed fixtures. • Runs real grasp selection under noise, vibration and production variability. • Maintains throughput while avoiding breakage on a delicate material. • Shows mobile manipulation doing actual shift work instead of controlled lab runs. Kinisi published a video that shows what the robot sees and how sensor data turns into action. This is the part most teams struggle to explain to customers, so the educational angle is useful for anyone working on adoption. On top of this, the team signed a pilot with a global automotive manufacturer to explore humanoid use cases in production. The direction is clear. Wheeled mobility (not legs!) plus strong perception seems to be shaping a large part of industrial humanoids right now. I know Brennand from earlier conversations and from our podcast session, and I am always glad to see European teams push the category forward. Wishing the Kinisi team continued success. —- Weekly robotics and AI insights. Subscribe free:
0:17

Most humanoid projects talk about real work. Very few last an hour on a real line. This week I saw a case that matters for anyone building robots, perception, or physical AI. Kinisi deployed its first mobile manipulation system into a live recycling facility. Not a demo. Not a staged test. A real production line with real output pressure. Why this matters if you want robotics to deliver real value on your floor: • Handles mixed glass with random poses and no fixed fixtures. • Runs real grasp selection under noise, vibration and production variability. • Maintains throughput while avoiding breakage on a delicate material. • Shows mobile manipulation doing actual shift work instead of controlled lab runs. Kinisi published a video that shows what the robot sees and how sensor data turns into action. This is the part most teams struggle to explain to customers, so the educational angle is useful for anyone working on adoption. On top of this, the team signed a pilot with a global automotive manufacturer to explore humanoid use cases in production. The direction is clear. Wheeled mobility (not legs!) plus strong perception seems to be shaping a large part of industrial humanoids right now. I know Brennand from earlier conversations and from our podcast session, and I am always glad to see European teams push the category forward. Wishing the Kinisi team continued success. —- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

24,743 views • 8 months ago

🚨 EARTH IS SPINNING SLIGHTLY FASTER AND FOR THE FIRST TIME, SCIENTISTS ARE SERIOUSLY DISCUSSING A “NEGATIVE LEAP SECOND”. For decades, Earth’s rotation has been gradually slowing, which is why we occasionally add a leap second to keep atomic clocks in sync with the planet. But since around 2018–2020, something unusual has happened: Earth has been rotating faster than expected. Several days in recent years (including June 29, 2022) were among the shortest ever recorded by atomic clocks. This acceleration has forced timekeepers to consider something that has never been done before: subtracting a second from official time (a negative leap second) instead of adding one. Why this matters: • A negative leap second would mean clocks skip a second (e.g., jumping from 23:59:58 straight to 00:00:00) • It would be the first time in history this has happened • It could affect GPS, financial systems, telecommunications, and any infrastructure that relies on precise timing • The leading theory links the speedup to changes in Earth’s liquid outer core, though melting ice sheets are partially counteracting the effect The deeper implication: Earth is not a perfect clock. Its rotation speed is influenced by complex interactions between the core, oceans, atmosphere, and even climate change. As we become more dependent on ultra-precise global timing, these small geophysical changes are no longer just scientific curiosities they’re potential infrastructure issues. A 2024 study in Nature suggested a negative leap second might be needed around 2029, though the exact timing remains uncertain and depends on how Earth’s rotation behaves in the coming years. Would you rather we keep adding/subtracting leap seconds forever, or should we eventually decouple our clocks from Earth’s imperfect rotation? Follow for more frontier Earth science and timekeeping realities.
0:10

🚨 EARTH IS SPINNING SLIGHTLY FASTER AND FOR THE FIRST TIME, SCIENTISTS ARE SERIOUSLY DISCUSSING A “NEGATIVE LEAP SECOND”. For decades, Earth’s rotation has been gradually slowing, which is why we occasionally add a leap second to keep atomic clocks in sync with the planet. But since around 2018–2020, something unusual has happened: Earth has been rotating faster than expected. Several days in recent years (including June 29, 2022) were among the shortest ever recorded by atomic clocks. This acceleration has forced timekeepers to consider something that has never been done before: subtracting a second from official time (a negative leap second) instead of adding one. Why this matters: • A negative leap second would mean clocks skip a second (e.g., jumping from 23:59:58 straight to 00:00:00) • It would be the first time in history this has happened • It could affect GPS, financial systems, telecommunications, and any infrastructure that relies on precise timing • The leading theory links the speedup to changes in Earth’s liquid outer core, though melting ice sheets are partially counteracting the effect The deeper implication: Earth is not a perfect clock. Its rotation speed is influenced by complex interactions between the core, oceans, atmosphere, and even climate change. As we become more dependent on ultra-precise global timing, these small geophysical changes are no longer just scientific curiosities they’re potential infrastructure issues. A 2024 study in Nature suggested a negative leap second might be needed around 2029, though the exact timing remains uncertain and depends on how Earth’s rotation behaves in the coming years. Would you rather we keep adding/subtracting leap seconds forever, or should we eventually decouple our clocks from Earth’s imperfect rotation? Follow for more frontier Earth science and timekeeping realities.

TheNewPhysics

48,968 views • 1 month ago

Yesterday at Brown University ICERM's workshop on “Agentic Scientific Computing and Scientific Machine Learning” I spoke about “Adaptive Swarms Across Scales”, making the case for scientific AI as systems that can create representations, stress them, fracture them, and enlarge the category in which future representations live. The category here is a composable and breakable working universe of science: data, hypotheses, simulations, measurements, tools, failures, figures, papers, provenance, and the transformations that connect them. Discovery happens when those transformations become executable, inspectable, composable, and capable of changing the world model they operate within. Atomistic modeling gives one category - states, forces, trajectories, observables, boundary conditions, conservation laws. Neural surrogates learn fast morphisms inside or between such categories. But discovery is higher-order: it changes which objects and morphisms are available in the first place: what variables exist, what operations are allowed, what evidence counts, what scale is active, what invariant is being preserved, and what kind of explanation the system is even capable of forming. This is scientific method as adaptive architecture: compression, stress, fracture, recomposition. Fracture matters here because it makes the logic physical: a non-commuting diagram realized in matter. The imposed load, material hierarchy, defect field, and assumed continuum description no longer map cleanly into the observed outcome. The crack is the obstruction and it identifies where the old morphism failed and where a new representation must be introduced. The physical crack and the categorical obstruction are the same event viewed in different substrates. ScienceClaw × Infinite is a machine for constructing and transforming a category of scientific artifacts. Each artifact is typed. Each operation has lineage. Each failed branch remains in the category as reusable structure. The “paper” is no longer the terminal object of science; it is one projection of a larger compositional trace, and it can be generated at any time for consumption by a human or an AI. With that the unit of scientific labor is changing. For most of the twentieth century the unit was the result (a measurement, a theorem, a synthesized molecule). It is now becoming the algorithm that produces results, and after that, the substrate of discovery itself. The static PDF is the wrong terminal object for this regime, and the role of the scientist with it. We now design algorithms that build algorithms, and eventually substrates in which such algorithms compose themselves. At that point, the scientist is no longer outside the discovery system. The scientist becomes one of the representations the system can transform. In that sense, the systems will eventually do science to us, and that is the structural consequence of the principle they are built on.
0:12

Yesterday at Brown University ICERM's workshop on “Agentic Scientific Computing and Scientific Machine Learning” I spoke about “Adaptive Swarms Across Scales”, making the case for scientific AI as systems that can create representations, stress them, fracture them, and enlarge the category in which future representations live. The category here is a composable and breakable working universe of science: data, hypotheses, simulations, measurements, tools, failures, figures, papers, provenance, and the transformations that connect them. Discovery happens when those transformations become executable, inspectable, composable, and capable of changing the world model they operate within. Atomistic modeling gives one category - states, forces, trajectories, observables, boundary conditions, conservation laws. Neural surrogates learn fast morphisms inside or between such categories. But discovery is higher-order: it changes which objects and morphisms are available in the first place: what variables exist, what operations are allowed, what evidence counts, what scale is active, what invariant is being preserved, and what kind of explanation the system is even capable of forming. This is scientific method as adaptive architecture: compression, stress, fracture, recomposition. Fracture matters here because it makes the logic physical: a non-commuting diagram realized in matter. The imposed load, material hierarchy, defect field, and assumed continuum description no longer map cleanly into the observed outcome. The crack is the obstruction and it identifies where the old morphism failed and where a new representation must be introduced. The physical crack and the categorical obstruction are the same event viewed in different substrates. ScienceClaw × Infinite is a machine for constructing and transforming a category of scientific artifacts. Each artifact is typed. Each operation has lineage. Each failed branch remains in the category as reusable structure. The “paper” is no longer the terminal object of science; it is one projection of a larger compositional trace, and it can be generated at any time for consumption by a human or an AI. With that the unit of scientific labor is changing. For most of the twentieth century the unit was the result (a measurement, a theorem, a synthesized molecule). It is now becoming the algorithm that produces results, and after that, the substrate of discovery itself. The static PDF is the wrong terminal object for this regime, and the role of the scientist with it. We now design algorithms that build algorithms, and eventually substrates in which such algorithms compose themselves. At that point, the scientist is no longer outside the discovery system. The scientist becomes one of the representations the system can transform. In that sense, the systems will eventually do science to us, and that is the structural consequence of the principle they are built on.

Markus J. Buehler

10,095 views • 2 months ago

Robots can now reconstruct 3D scenes in real time from a single RGB camera. [📍 Projects page + paper] No depth sensor. No retraining. 30 FPS. Researchers at the Imperial College London introduced KV-Tracker, a training-free method that makes heavy models like π³ and Depth Anything 3 fast enough for real-time tracking. The idea is simple. These models use global self-attention, which is powerful but computationally expensive. KV-Tracker caches the key and value pairs from selected keyframes and reuses them for new frames. That cache becomes an implicit scene representation. Result: • Up to 30 FPS • 10 to 15x speedup • Accurate 6-DoF tracking on benchmarks like TUM RGB-D and 7-Scenes • Works with monocular RGB only It also supports object-level tracking with masks and allows saving the KV-cache for later reuse. For robotics, this reduces hardware constraints and moves real-time 3D perception closer to practical deployment. Credit to Marwan Taher (Marwan Taher) at Imperial’s Dyson Robotics Lab and many others who contributed to this! 📍 Save projects page + paper for later: Video: ——- if it matters in AI or Robotics you'll read it here first:
0:18

Robots can now reconstruct 3D scenes in real time from a single RGB camera. [📍 Projects page + paper] No depth sensor. No retraining. 30 FPS. Researchers at the Imperial College London introduced KV-Tracker, a training-free method that makes heavy models like π³ and Depth Anything 3 fast enough for real-time tracking. The idea is simple. These models use global self-attention, which is powerful but computationally expensive. KV-Tracker caches the key and value pairs from selected keyframes and reuses them for new frames. That cache becomes an implicit scene representation. Result: • Up to 30 FPS • 10 to 15x speedup • Accurate 6-DoF tracking on benchmarks like TUM RGB-D and 7-Scenes • Works with monocular RGB only It also supports object-level tracking with masks and allows saving the KV-cache for later reuse. For robotics, this reduces hardware constraints and moves real-time 3D perception closer to practical deployment. Credit to Marwan Taher (Marwan Taher) at Imperial’s Dyson Robotics Lab and many others who contributed to this! 📍 Save projects page + paper for later: Video: ——- if it matters in AI or Robotics you'll read it here first:

Ilir Aliu

53,911 views • 4 months ago

Most video tools can generate clips. Very few can maintain identity. That has been the real bottleneck in AI video creation. Kling O1 changes that. For the first time, creators can carry a character, style, and visual language across scenes without constant fixes. You can reference past clips, assets, or images and the output stays consistently on-model. No visual drift. No rework loops. No “this doesn’t look like the last shot” moments. It feels less like prompting a tool and more like working with a creative collaborator that remembers context. The impact is practical, not theoretical: → Faster production cycles → Lower iteration costs → Noticeably higher output quality This is what mature AI tooling looks like. Not louder features. Not bigger claims. Just reliability where it actually matters. Consistency is no longer the problem.
0:13

Most video tools can generate clips. Very few can maintain identity. That has been the real bottleneck in AI video creation. Kling O1 changes that. For the first time, creators can carry a character, style, and visual language across scenes without constant fixes. You can reference past clips, assets, or images and the output stays consistently on-model. No visual drift. No rework loops. No “this doesn’t look like the last shot” moments. It feels less like prompting a tool and more like working with a creative collaborator that remembers context. The impact is practical, not theoretical: → Faster production cycles → Lower iteration costs → Noticeably higher output quality This is what mature AI tooling looks like. Not louder features. Not bigger claims. Just reliability where it actually matters. Consistency is no longer the problem.

Darshal Jaitwar

141,038 views • 7 months ago

Trained on zero real-world data. Learned to walk, pick up boxes, and follow multi-step instructions... in the REAL world. ( 📌 Paper below) Researchers from Amazon FAR, Berkeley, Stanford, and CMU scanned real rooms with an iPhone, rebuilt them as 3D Gaussian Splatting scenes, then generated 48,000 synthetic trajectories of a Unitree G1 walking, grasping, and placing objects inside those virtual replicas. They rendered the robot's first-person camera view from each run and paired it with the matching language instruction and motion data. That's the dataset every humanoid team needs and nobody has: synced egocentric video + language + kinematics, at scale. Instead of collecting it in the real world, they manufactured it. They trained a vision-language-kinematics policy on that synthetic data alone, then deployed it on the physical G1 across five task types: navigation to a named object, lifting boxes of three different sizes with no per-size tuning, chained multi-step tasks, robustness to mid-task layout changes and flickering lights, and multi-minute long-horizon runs. No real-world fine-tuning at any point. Real-world interaction data has been the hard limit on humanoid learning... slow, expensive, and small. If scanning a room once and synthesizing thousands of labeled interactions holds up as a general recipe, that limit moves. Data stops being the bottleneck robotics teams have to solve for. 📌 Paper: Project: ——- Weekly robotics and AI insights. Subscribe free:
0:15

Trained on zero real-world data. Learned to walk, pick up boxes, and follow multi-step instructions... in the REAL world. ( 📌 Paper below) Researchers from Amazon FAR, Berkeley, Stanford, and CMU scanned real rooms with an iPhone, rebuilt them as 3D Gaussian Splatting scenes, then generated 48,000 synthetic trajectories of a Unitree G1 walking, grasping, and placing objects inside those virtual replicas. They rendered the robot's first-person camera view from each run and paired it with the matching language instruction and motion data. That's the dataset every humanoid team needs and nobody has: synced egocentric video + language + kinematics, at scale. Instead of collecting it in the real world, they manufactured it. They trained a vision-language-kinematics policy on that synthetic data alone, then deployed it on the physical G1 across five task types: navigation to a named object, lifting boxes of three different sizes with no per-size tuning, chained multi-step tasks, robustness to mid-task layout changes and flickering lights, and multi-minute long-horizon runs. No real-world fine-tuning at any point. Real-world interaction data has been the hard limit on humanoid learning... slow, expensive, and small. If scanning a room once and synthesizing thousands of labeled interactions holds up as a general recipe, that limit moves. Data stops being the bottleneck robotics teams have to solve for. 📌 Paper: Project: ——- Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

12,950 views • 16 days ago

I spent a month in Shenzhen visiting factories and robotics companies, and the contrast with the U.S. was striking. While Figure and Boston Dynamics hide their humanoids behind closed doors, Chinese companies have massive showrooms open to the public. But what really stood out wasn't just the transparency, it was how good they are at selling. Take UBTech: they've already sold 1,200 humanoid units at $200k each to factories. And here's the kicker, these robots aren't even that useful yet. They can only pick up and drop boxes at 1/10th the speed of a human, and factories still need to hire system integrators to train them for specific tasks. My theory is that these factories are terrified of getting left behind in the robotics/AI wave. They're investing in new tech not because it's ready, but because they can't afford to wait. The second surprise was the breadth of their robotics portfolio. These companies aren't just building humanoids, they're deploying service robots everywhere: restaurants, hotels, apartments. Consumer robots are cleaning houses, pools, pet waste, dishes. They're covering the entire spectrum. But the education piece shocked me most. I picked up what I thought was a high school or college robotics textbook, it was for primary school. The government mandated AI and robotics education starting in elementary school. Almost every single school in China now has AI and robotics curriculum, complete with education robots so kids can learn by building. They're creating a generation that grows up fluent in robotics and AI. China owns the supply chain and the hardware stack. But here's what I think people are missing: the race isn't just about who can build robots faster or cheaper. The U.S. advantage has always been in the layer between hardware and human, the interaction design, the software intelligence, the intuitive interfaces that make complex technology feel natural. China is building the physical infrastructure, but they're also learning fast. Every deployed service robot, every classroom full of kids building with education kits, every factory running humanoids, that's all data collection at scale. The window for the U.S. to establish its wedge is narrowing. It's not enough to be better at AI or software anymore. We need to be building the integration layer, the intelligence that makes physical AI actually useful, not just impressive in a showroom. Because right now, China isn't just manufacturing robots. They're manufacturing a robotics-native culture, and that might be the most defensible moat of all.
0:10

I spent a month in Shenzhen visiting factories and robotics companies, and the contrast with the U.S. was striking. While Figure and Boston Dynamics hide their humanoids behind closed doors, Chinese companies have massive showrooms open to the public. But what really stood out wasn't just the transparency, it was how good they are at selling. Take UBTech: they've already sold 1,200 humanoid units at $200k each to factories. And here's the kicker, these robots aren't even that useful yet. They can only pick up and drop boxes at 1/10th the speed of a human, and factories still need to hire system integrators to train them for specific tasks. My theory is that these factories are terrified of getting left behind in the robotics/AI wave. They're investing in new tech not because it's ready, but because they can't afford to wait. The second surprise was the breadth of their robotics portfolio. These companies aren't just building humanoids, they're deploying service robots everywhere: restaurants, hotels, apartments. Consumer robots are cleaning houses, pools, pet waste, dishes. They're covering the entire spectrum. But the education piece shocked me most. I picked up what I thought was a high school or college robotics textbook, it was for primary school. The government mandated AI and robotics education starting in elementary school. Almost every single school in China now has AI and robotics curriculum, complete with education robots so kids can learn by building. They're creating a generation that grows up fluent in robotics and AI. China owns the supply chain and the hardware stack. But here's what I think people are missing: the race isn't just about who can build robots faster or cheaper. The U.S. advantage has always been in the layer between hardware and human, the interaction design, the software intelligence, the intuitive interfaces that make complex technology feel natural. China is building the physical infrastructure, but they're also learning fast. Every deployed service robot, every classroom full of kids building with education kits, every factory running humanoids, that's all data collection at scale. The window for the U.S. to establish its wedge is narrowing. It's not enough to be better at AI or software anymore. We need to be building the integration layer, the intelligence that makes physical AI actually useful, not just impressive in a showroom. Because right now, China isn't just manufacturing robots. They're manufacturing a robotics-native culture, and that might be the most defensible moat of all.

Miyu Horiuchi

90,718 views • 6 months ago

Elon Musk just identified the next crisis in AI. It’s not a shortage. It’s an unusable surplus. Musk: “By the end of this year, chip production will outpace the ability to turn chips on.” For three years the world was starved for silicon. Every lab, every government, every company racing to secure the chips that determine who wins the AI era. That bottleneck is ending. A new one is replacing it. Musk: “The chips are going to be piling up and not be able to be turned on.” Billions of dollars of the most advanced AI hardware ever built. Sitting dark. Not because the chips don’t work. Because there isn’t enough electricity to run them. You can’t print a power plant the way you print a chip. The fabrication plants scaled. The grid didn’t. And now the most valuable hardware in history is about to hit a wall that no amount of capital can instantly solve. Compute is about to become abundant. Electricity is about to become the most valuable commodity on earth. Three years obsessing over silicon yields. Physics doesn’t care about your chip architecture if your data center can’t pull enough megawatts. The war isn’t about who can manufacture the most silicon anymore. It’s about who has the raw power to plug it in. Whoever solves energy first doesn’t just win. They own the infrastructure everyone else needs to compete. The losers stack useless chips in warehouses waiting for power that never arrives. We built a trillion dollar engine and forgot the fuel. That’s the AI race right now.
0:10

Elon Musk just identified the next crisis in AI. It’s not a shortage. It’s an unusable surplus. Musk: “By the end of this year, chip production will outpace the ability to turn chips on.” For three years the world was starved for silicon. Every lab, every government, every company racing to secure the chips that determine who wins the AI era. That bottleneck is ending. A new one is replacing it. Musk: “The chips are going to be piling up and not be able to be turned on.” Billions of dollars of the most advanced AI hardware ever built. Sitting dark. Not because the chips don’t work. Because there isn’t enough electricity to run them. You can’t print a power plant the way you print a chip. The fabrication plants scaled. The grid didn’t. And now the most valuable hardware in history is about to hit a wall that no amount of capital can instantly solve. Compute is about to become abundant. Electricity is about to become the most valuable commodity on earth. Three years obsessing over silicon yields. Physics doesn’t care about your chip architecture if your data center can’t pull enough megawatts. The war isn’t about who can manufacture the most silicon anymore. It’s about who has the raw power to plug it in. Whoever solves energy first doesn’t just win. They own the infrastructure everyone else needs to compete. The losers stack useless chips in warehouses waiting for power that never arrives. We built a trillion dollar engine and forgot the fuel. That’s the AI race right now.

Dustin

705,429 views • 5 months ago

Model-Free Reinforcement Learning (MFRL) has been alluring, especially with supercharged compute with physics on GPU. However, the methods use 0-th order gradients, and are often not the best optimizers. Can we do better than PPO in continuous control for robotics? Turns out yes! 🥳 tl;dr: Faster, better RL than PPO in continuous control 💪 The answer lies in using more information from the simulation. We are juicing the simulation on GPU as it is, why not use it for gradients as well? This has been a driving question in a series of our works. We first studied this problem in ICLR 2022 paper on Short Horizon Actor Critic Naive gradient based methods are stuck in local minima and have exploding/vanishing gradients. SHAC solved this problem truncated rollouts and model based value estimation, where the model is Differentiable Sim. This boosted sample efficiency and wall-clock time immensely especially in high dimensional systems such as humanoids Yet, given enough compute PPO often caught up. Our follow up paper on on Adaptive Horizon Actor Critic at ICML 2024 discovers the cause and provides a fix. However, we find that even when given ground-truth dynamics, not all gradients are useful due to sample error. 1st-Order Model-Based Reinforcement Learning methods employing differentiable simulation provide gradients with reduced variance but are susceptible to bias in scenarios involving stiff dynamics, such as physical contact. We find that back-propagating through contact and long trajectories drastically reduces gradient accuracy. Using this insight, we propose AHAC to dynamically adapt its roll-out horizon to avoid differentiating through stiff contact. AHAC is a first-order model-based RL algorithm that learns high-dimensional tasks in minutes (wall clock) and outperforms PPO by 40%, even in the limit of data provided to PPO. This work is led by Ignat Georgiev alongside Krishnan Srinivasan, Jie Xu, Eric Heiden and ample assistance from warp team at NVIDIA Robotics (Miles Macklin)
0:10

Model-Free Reinforcement Learning (MFRL) has been alluring, especially with supercharged compute with physics on GPU. However, the methods use 0-th order gradients, and are often not the best optimizers. Can we do better than PPO in continuous control for robotics? Turns out yes! 🥳 tl;dr: Faster, better RL than PPO in continuous control 💪 The answer lies in using more information from the simulation. We are juicing the simulation on GPU as it is, why not use it for gradients as well? This has been a driving question in a series of our works. We first studied this problem in ICLR 2022 paper on Short Horizon Actor Critic Naive gradient based methods are stuck in local minima and have exploding/vanishing gradients. SHAC solved this problem truncated rollouts and model based value estimation, where the model is Differentiable Sim. This boosted sample efficiency and wall-clock time immensely especially in high dimensional systems such as humanoids Yet, given enough compute PPO often caught up. Our follow up paper on on Adaptive Horizon Actor Critic at ICML 2024 discovers the cause and provides a fix. However, we find that even when given ground-truth dynamics, not all gradients are useful due to sample error. 1st-Order Model-Based Reinforcement Learning methods employing differentiable simulation provide gradients with reduced variance but are susceptible to bias in scenarios involving stiff dynamics, such as physical contact. We find that back-propagating through contact and long trajectories drastically reduces gradient accuracy. Using this insight, we propose AHAC to dynamically adapt its roll-out horizon to avoid differentiating through stiff contact. AHAC is a first-order model-based RL algorithm that learns high-dimensional tasks in minutes (wall clock) and outperforms PPO by 40%, even in the limit of data provided to PPO. This work is led by Ignat Georgiev alongside Krishnan Srinivasan, Jie Xu, Eric Heiden and ample assistance from warp team at NVIDIA Robotics (Miles Macklin)

Animesh Garg

52,308 views • 2 years ago

Google 3D Maps aren’t as accurate as you think. Everyone knows Google dominates mapping—but here’s what they won’t tell you. Their 3D maps are built with car-mounted cameras, capturing images every ~10 meters. That’s fine for basic navigation, but it’s nowhere near enough for AR, robotics, autonomous systems, or AI-driven spatial intelligence. Now, imagine a map so detailed it captures the world at sub-5cm accuracy. That’s OVRMaps. 🌐 OVER 3D maps are built with 400-1,000 images per 300 sqm, taken from multiple angles at pedestrian level. The result? A new era of hyper-precise localization that changes everything. Why does this matter? ↳ VPS that actually works – Real-world AR anchoring, digital twins, and AI-powered spatial computing with pinpoint accuracy. ↳ Next-gen precision – Essential for ride-sharing, robotics, smart cities, and asset tracking—where even a small error makes a huge difference. ↳ AI-Driven Spatial Intelligence – Our maps fuel Large Geo-Spatial Models (LGMs), the AI revolution enabling machines to understand, navigate, and interact with the real world. ↳ The Spatial Computing revolution – Merging immersive experiences with real-world precision, unlocking applications beyond what Google’s dataset can support. OVRMaps are built for the future. 🌍 Start mapping today:
0:12

Google 3D Maps aren’t as accurate as you think. Everyone knows Google dominates mapping—but here’s what they won’t tell you. Their 3D maps are built with car-mounted cameras, capturing images every ~10 meters. That’s fine for basic navigation, but it’s nowhere near enough for AR, robotics, autonomous systems, or AI-driven spatial intelligence. Now, imagine a map so detailed it captures the world at sub-5cm accuracy. That’s OVRMaps. 🌐 OVER 3D maps are built with 400-1,000 images per 300 sqm, taken from multiple angles at pedestrian level. The result? A new era of hyper-precise localization that changes everything. Why does this matter? ↳ VPS that actually works – Real-world AR anchoring, digital twins, and AI-powered spatial computing with pinpoint accuracy. ↳ Next-gen precision – Essential for ride-sharing, robotics, smart cities, and asset tracking—where even a small error makes a huge difference. ↳ AI-Driven Spatial Intelligence – Our maps fuel Large Geo-Spatial Models (LGMs), the AI revolution enabling machines to understand, navigate, and interact with the real world. ↳ The Spatial Computing revolution – Merging immersive experiences with real-world precision, unlocking applications beyond what Google’s dataset can support. OVRMaps are built for the future. 🌍 Start mapping today:

Over the Reality 🌐

1,331,567 views • 1 year ago

When CK was kneeling, that was a defining moment for the culture, I believe it created a rare moment of leverage for blacks, one of the first since I believe since Martin Luther King Jr. where the Black community was visibly unified and the country was forced to look. We were together, and it threatened the power. Then Jay-Z, with all his influence on the culture stepped in as a bridge for the establishment, and the momentum collapsed. Just like that, the pressure was gone. Our unity was neutralized. And that’s what that mfer took from us. Now fast-forward. You see Jay-Z seated at elite tables, surrounded by power and exclusivity. And it’s hard not to read that moment as the reward for selling his people out. That’s not legacy, that’s payment. That’s the prize for defusing a movement that could’ve forced real recognition and real leverage for Black people as a collective. That seat didn’t come from lifting us, it came from standing on a movement we created and redirecting it to them. We are observing incentives⬇️. Influence like that doesn’t come free, and loyalty that unanimous doesn’t happen by accident. When every major voice goes quiet at the same time, when criticism disappears overnight, it raises a simple question: who benefited, and who paid the cost? That dinner wasn’t symbolism, I bet. It was confirmation. The moment was traded, and the bill was paid by us. Power always selects a familiar face. Not to free us but to manage us. Kendrick isn’t being elevated by accident. He’s being positioned another puppet. Strip away all the rhetoric, and look at the incentives. That’s how manipulation works. Idiots. Wake up.
0:16

When CK was kneeling, that was a defining moment for the culture, I believe it created a rare moment of leverage for blacks, one of the first since I believe since Martin Luther King Jr. where the Black community was visibly unified and the country was forced to look. We were together, and it threatened the power. Then Jay-Z, with all his influence on the culture stepped in as a bridge for the establishment, and the momentum collapsed. Just like that, the pressure was gone. Our unity was neutralized. And that’s what that mfer took from us. Now fast-forward. You see Jay-Z seated at elite tables, surrounded by power and exclusivity. And it’s hard not to read that moment as the reward for selling his people out. That’s not legacy, that’s payment. That’s the prize for defusing a movement that could’ve forced real recognition and real leverage for Black people as a collective. That seat didn’t come from lifting us, it came from standing on a movement we created and redirecting it to them. We are observing incentives⬇️. Influence like that doesn’t come free, and loyalty that unanimous doesn’t happen by accident. When every major voice goes quiet at the same time, when criticism disappears overnight, it raises a simple question: who benefited, and who paid the cost? That dinner wasn’t symbolism, I bet. It was confirmation. The moment was traded, and the bill was paid by us. Power always selects a familiar face. Not to free us but to manage us. Kendrick isn’t being elevated by accident. He’s being positioned another puppet. Strip away all the rhetoric, and look at the incentives. That’s how manipulation works. Idiots. Wake up.

industrypolitics

25,074 views • 6 months ago

A new era for Nodepay has begun. Bandwidth has been powering our real-time data retrieval infrastructure from day one. Now, that foundation has delivered something bigger: The Intelligence Engine — a living network of structured, real-time insights, powered by you. - Your predictions become real-time signals - Your insights become knowledge - Your bandwidth collects data for our AI - $NC becomes the fuel This is AI combined with collective intelligence. Why It Matters: - Users get rewarded for their insights, not just engagement - Projects gain direct access to real-time, validated community sentiment - Everyone saves time, earns more, & makes smarter decisions First access is reserved for $NC holders ⚠️ The next wave of users will be onboarded following the upcoming snapshot 📸 Soon, you’ll all be able to contribute insights, unlock smarter rewards, and tap into real-time sentiment and conviction that gives you the edge. The Intelligence Engine is live — and you’re getting paid to power it.
0:10

A new era for Nodepay has begun. Bandwidth has been powering our real-time data retrieval infrastructure from day one. Now, that foundation has delivered something bigger: The Intelligence Engine — a living network of structured, real-time insights, powered by you. - Your predictions become real-time signals - Your insights become knowledge - Your bandwidth collects data for our AI - $NC becomes the fuel This is AI combined with collective intelligence. Why It Matters: - Users get rewarded for their insights, not just engagement - Projects gain direct access to real-time, validated community sentiment - Everyone saves time, earns more, & makes smarter decisions First access is reserved for $NC holders ⚠️ The next wave of users will be onboarded following the upcoming snapshot 📸 Soon, you’ll all be able to contribute insights, unlock smarter rewards, and tap into real-time sentiment and conviction that gives you the edge. The Intelligence Engine is live — and you’re getting paid to power it.

Nodepay

82,903 views • 1 year ago