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。

this is f*cking gold - I built a neural-vision prototype for humanoid robotics. With GPT Astra, my Robinhood token experiment took a very unexpected turn. CatBrain was originally meant ONLY for Robinhood tokens. Then I started experimenting beyond markets. → Cat-inspired visual processing: ON/OFF contrast channels and motion signals... show more

BuBBliK

23,800 subscribers

38,843 次观看 • 8 天前 •via X (Twitter)

艺术教育科学技术

Anya Rossi• Live Now

Private livecam show

2 条评论

rewind 的头像

rewind8 天前

Unitree H1 makes sense

Ibriham A Turay 的头像

Ibriham A Turay8 天前

Explain this cat brain for me ?

相关视频

this is insane - I made a robot that plays ping-pong using CAT-INSPIRED vision A camera watches the ball. 20,480 artificial neurons process what it sees. Then a mechanical arm predicts where the ball is going - and moves to intercept it. Here’s what’s happening under the hood: → 20,480 ARTIFICIAL NEURONS process visual signals using a leaky integrate-and-fire model → 8 ORIENTATION CHANNELS detect edges and structure inspired by mechanisms studied in feline vision → The system tracks the ball and predicts where it will be around 80 MS into the future. → The FULL pipeline now works: camera → neural activity → tracking → trajectory → robotic arm movement What started as a weird experiment is now a WORKING physical system. A cat-inspired visual model is literally controlling a robot arm to chase a ping-pong ball. GitHub is in the reply. BOOKMARK this - I’ll keep posting updates as I push catbrain further.

this is insane - I made a robot that plays ping-pong using CAT-INSPIRED vision A camera watches the ball. 20,480 artificial neurons process what it sees. Then a mechanical arm predicts where the ball is going - and moves to intercept it. Here’s what’s happening under the hood: → 20,480 ARTIFICIAL NEURONS process visual signals using a leaky integrate-and-fire model → 8 ORIENTATION CHANNELS detect edges and structure inspired by mechanisms studied in feline vision → The system tracks the ball and predicts where it will be around 80 MS into the future. → The FULL pipeline now works: camera → neural activity → tracking → trajectory → robotic arm movement What started as a weird experiment is now a WORKING physical system. A cat-inspired visual model is literally controlling a robot arm to chase a ping-pong ball. GitHub is in the reply. BOOKMARK this - I’ll keep posting updates as I push catbrain further.

BuBBliK

14,056 次观看 • 6 天前

I built GTAFLY with GPT-6 Astra A simulated fly brain that drives a car in GTA V. The game's visual feed is converted into neural signals and processed across 166,700 simulated neurons. Their activity controls steering, acceleration, and braking, while dopamine-like rewards reinforce successful movement and mistakes reshape future responses. Inspired by Stonkfly. I added the GitHub repository with the code in the comments. Now I’m curious whether GTAFLY could be transferred to GTA VI.

I built GTAFLY with GPT-6 Astra A simulated fly brain that drives a car in GTA V. The game's visual feed is converted into neural signals and processed across 166,700 simulated neurons. Their activity controls steering, acceleration, and braking, while dopamine-like rewards reinforce successful movement and mistakes reshape future responses. Inspired by Stonkfly. I added the GitHub repository with the code in the comments. Now I’m curious whether GTAFLY could be transferred to GTA VI.

BuBBliK

41,942 次观看 • 12 天前

$BRAIN is live. the open-source fly-brain bridge has a ticker now, and the ticker is becoming the input for a physical robot. CA: 9LuwgFQemAoV9rgVBBtwbRSxBmamcRRbysEK8yL2pump what’s already done: the neural bridge is working. camera input becomes activity across eight virtual neural populations, and that activity becomes left and right motor commands IMU data feeds movement back into the model. smoothing, speed limits and a 500 ms watchdog are already built in the synthetic demo runs locally today. the ESP32 scaffold is ready for hardware integration the entire project is open source and MIT licensed github: what’s next, in order: the on-chain listener goes live. every $BRAIN transaction becomes a stimulus sent directly into the neural model then the physical robot connects. the wallet address determines which neurons activate, the amount determines the strength of the impulse, and the brain converts that reaction into movement then the livestream. every transaction, neural impulse and physical response visible in real time then user commands. spend $BRAIN to trigger a stronger reflex, request movement, temporarily control the robot, name a neuron or place your name on the stream the loop, plain: a transaction enters the neurons react the robot moves a Proof of Reflex is created with the transaction hash, neural activity map and video of the physical response the current brain is a small fly-inspired simulation. the bridge and local demo work today. the blockchain connection, physical robot and livestream are being built now the internet becomes its sensory organ. the blockchain becomes its nervous system. $BRAIN makes the body move.

$BRAIN is live. the open-source fly-brain bridge has a ticker now, and the ticker is becoming the input for a physical robot. CA: 9LuwgFQemAoV9rgVBBtwbRSxBmamcRRbysEK8yL2pump what’s already done: the neural bridge is working. camera input becomes activity across eight virtual neural populations, and that activity becomes left and right motor commands IMU data feeds movement back into the model. smoothing, speed limits and a 500 ms watchdog are already built in the synthetic demo runs locally today. the ESP32 scaffold is ready for hardware integration the entire project is open source and MIT licensed github: what’s next, in order: the on-chain listener goes live. every $BRAIN transaction becomes a stimulus sent directly into the neural model then the physical robot connects. the wallet address determines which neurons activate, the amount determines the strength of the impulse, and the brain converts that reaction into movement then the livestream. every transaction, neural impulse and physical response visible in real time then user commands. spend $BRAIN to trigger a stronger reflex, request movement, temporarily control the robot, name a neuron or place your name on the stream the loop, plain: a transaction enters the neurons react the robot moves a Proof of Reflex is created with the transaction hash, neural activity map and video of the physical response the current brain is a small fly-inspired simulation. the bridge and local demo work today. the blockchain connection, physical robot and livestream are being built now the internet becomes its sensory organ. the blockchain becomes its nervous system. $BRAIN makes the body move.

Frank

342,545 次观看 • 9 天前

I uploaded a fly brain and GPT-6 Astra into a Vector robot I bought on eBay Vector was supposed to be a desk pet. Now it has a reconstructed fruit fly nervous system running inside it. MaleCNS gives it 166,700 simulated neurons and 25.6 million connections. GPT-6 Astra gives it vision. Astra sees the room through Vector's camera - furniture, walls, doors, objects on the floor - and turns the visual scene into sensory signals. The fly brain processes those signals and chooses where Vector should go next. Astra sees. The fly brain decides. Vector moves. > No joystick > No preset route > No one manually telling it to turn left or right It started roaming around my apartment, stopping in front of furniture, changing direction, and reacting to things in its path. I bought it on eBay because it looked like a cute little WALL-E. Now it has a brain, eyes, and a body. And it keeps looking at me like it knows I’m the reason it exists.

I uploaded a fly brain and GPT-6 Astra into a Vector robot I bought on eBay Vector was supposed to be a desk pet. Now it has a reconstructed fruit fly nervous system running inside it. MaleCNS gives it 166,700 simulated neurons and 25.6 million connections. GPT-6 Astra gives it vision. Astra sees the room through Vector's camera - furniture, walls, doors, objects on the floor - and turns the visual scene into sensory signals. The fly brain processes those signals and chooses where Vector should go next. Astra sees. The fly brain decides. Vector moves. > No joystick > No preset route > No one manually telling it to turn left or right It started roaming around my apartment, stopping in front of furniture, changing direction, and reacting to things in its path. I bought it on eBay because it looked like a cute little WALL-E. Now it has a brain, eyes, and a body. And it keeps looking at me like it knows I’m the reason it exists.

kiruwaaaa

53,923 次观看 • 5 天前

ELON MUSK ON THE BRAIN’S INSANE VISION POST-PROCESSING ELON: “The brain is doing a crazy amount of post-processing on the vision signals from your eyes. It’s insane. Once you get all those vision signals, your brain is constantly trying to forget as much as possible so human memory is, perhaps the weakest thing about the brain is memory.” This exchange emphasizes the brain’s rigorous compression of visual data into essential vectors, mirroring efficiency in neural networks for real-time decision-making. The deeper meaning reveals cognition’s evolutionary prioritization of relevance over raw storage, informing scalable AI architectures. The technical ‘aha’ emerges in recognizing memory constraints as a catalyst for perceptual distillation, pivotal for advancing autonomous systems. Source: Elon Musk x Lex Fridman, December 28, 2021 — Topic: AI and Autonomous Driving.

ELON MUSK ON THE BRAIN’S INSANE VISION POST-PROCESSING ELON: “The brain is doing a crazy amount of post-processing on the vision signals from your eyes. It’s insane. Once you get all those vision signals, your brain is constantly trying to forget as much as possible so human memory is, perhaps the weakest thing about the brain is memory.” This exchange emphasizes the brain’s rigorous compression of visual data into essential vectors, mirroring efficiency in neural networks for real-time decision-making. The deeper meaning reveals cognition’s evolutionary prioritization of relevance over raw storage, informing scalable AI architectures. The technical ‘aha’ emerges in recognizing memory constraints as a catalyst for perceptual distillation, pivotal for advancing autonomous systems. Source: Elon Musk x Lex Fridman, December 28, 2021 — Topic: AI and Autonomous Driving.

Lacey

125,087 次观看 • 9 个月前

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Steve (Builder.io)

126,165 次观看 • 3 个月前

this is insane - flybrain beat elden ring I connected a fly's brain to Elden Ring, and five days later it had completed the entire main story, bosses included. GPT Astra helped me put the setup together in just a few prompts. → Game observations were encoded into sensory signals → The fly brain processed those signals into motor responses → Motor outputs triggered keyboard presses → Every resulting change fed back into the brain, starting the next cycle The strangest part was watching something built for a completely different world face Malenia through a keyboard. Code and tests are on GitHub - link in the comments.

this is insane - flybrain beat elden ring I connected a fly's brain to Elden Ring, and five days later it had completed the entire main story, bosses included. GPT Astra helped me put the setup together in just a few prompts. → Game observations were encoded into sensory signals → The fly brain processed those signals into motor responses → Motor outputs triggered keyboard presses → Every resulting change fed back into the brain, starting the next cycle The strangest part was watching something built for a completely different world face Malenia through a keyboard. Code and tests are on GitHub - link in the comments.

BuBBliK

44,952 次观看 • 10 天前

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Jim Fan

467,237 次观看 • 1 年前

$FOMOBRAIN is live. the neural copy of 169 winning memecoin traders on Robinhood has a ticker now, and the ticker is how the brain pays for itself. CA: 0x5cdf61bef103d9b9fffe2b6edb6aab541ceec1ac what's already done: - 116,420 fills, 5,918 tokens, 374 traders, 36 days of tape, all of it read straight off the chain - the brain itself. a neural copy of the 169 wallets that keep winning, built on GPT-6 Astra - live. bursts, fresh launches, exits, a score on every trader with the reasoning shown - 24/7 watch. the chain gets read every 20 seconds, every fill gets checked, nobody touches it - the whole code on github. 17,476 lines, MIT what's next, in order: - TG bot goes public. signals straight from the brain, in your pocket - a public trading algo on a real balance. public wallet, every trade readable on chain, the brain picks and i don't - the flywheel the flywheel, plain: signals in the bot are sold for $FOMOBRAIN. every token spent on them is burned every trade of $FOMOBRAIN pays creator fees. those fees are the algo's trading deposit, and the algo trades it on the brain's own signals 50% of what the algo makes goes back into the deposit. the other 50% buys $FOMOBRAIN off the market and burns it the finished algo gets sold to a closed group for $FOMOBRAIN, or rented for it. burned either way so every road ends at the same place. more subscribers, more burn. more volume, bigger deposit, more profit, bigger buyback, more burn no numbers promised. i don't know how fast this spins yet, i know which way it spins god bless

$FOMOBRAIN is live. the neural copy of 169 winning memecoin traders on Robinhood has a ticker now, and the ticker is how the brain pays for itself. CA: 0x5cdf61bef103d9b9fffe2b6edb6aab541ceec1ac what's already done: - 116,420 fills, 5,918 tokens, 374 traders, 36 days of tape, all of it read straight off the chain - the brain itself. a neural copy of the 169 wallets that keep winning, built on GPT-6 Astra - live. bursts, fresh launches, exits, a score on every trader with the reasoning shown - 24/7 watch. the chain gets read every 20 seconds, every fill gets checked, nobody touches it - the whole code on github. 17,476 lines, MIT what's next, in order: - TG bot goes public. signals straight from the brain, in your pocket - a public trading algo on a real balance. public wallet, every trade readable on chain, the brain picks and i don't - the flywheel the flywheel, plain: signals in the bot are sold for $FOMOBRAIN. every token spent on them is burned every trade of $FOMOBRAIN pays creator fees. those fees are the algo's trading deposit, and the algo trades it on the brain's own signals 50% of what the algo makes goes back into the deposit. the other 50% buys $FOMOBRAIN off the market and burns it the finished algo gets sold to a closed group for $FOMOBRAIN, or rented for it. burned either way so every road ends at the same place. more subscribers, more burn. more volume, bigger deposit, more profit, bigger buyback, more burn no numbers promised. i don't know how fast this spins yet, i know which way it spins god bless

cvxv666

154,565 次观看 • 11 天前

A DEAD FLY IS DRIVING MY ROBOT RIGHT NOW. Not a metaphor. A reconstructed neural map of a fruit fly - 166,700 neurons, 25.6 million synapses - is now running on my MacBook and steering a quadruped robot in my living room. Here's how it works: 1. MaleCNS - a full connectome reconstruction of an adult fly brain - runs as a live simulation on my laptop. 2. I used Codex to build a decoder that reads the simulation's neural activity and turns it into movement commands: forward, left, right. 3. Those commands get sent over Wi-Fi to the robot, which still uses its own firmware to drive all 12 servos. The brain only decides where to go, not how to move each leg. No scripts. No prerecorded paths. The simulation runs continuously, live, while the robot walks. The fly has been dead for years. But its reconstructed nervous system is producing real decisions again, in real time, controlling a machine that didn't exist when it was alive. Next step: feed the robot's sensors back into the simulation and close the loop. A dead brain that can finally feel the world it's walking through. This fly has been dead for years. Its brain runs on my laptop. Its decisions move a machine through my living room. One dead nervous system, walking again.

A DEAD FLY IS DRIVING MY ROBOT RIGHT NOW. Not a metaphor. A reconstructed neural map of a fruit fly - 166,700 neurons, 25.6 million synapses - is now running on my MacBook and steering a quadruped robot in my living room. Here's how it works: 1. MaleCNS - a full connectome reconstruction of an adult fly brain - runs as a live simulation on my laptop. 2. I used Codex to build a decoder that reads the simulation's neural activity and turns it into movement commands: forward, left, right. 3. Those commands get sent over Wi-Fi to the robot, which still uses its own firmware to drive all 12 servos. The brain only decides where to go, not how to move each leg. No scripts. No prerecorded paths. The simulation runs continuously, live, while the robot walks. The fly has been dead for years. But its reconstructed nervous system is producing real decisions again, in real time, controlling a machine that didn't exist when it was alive. Next step: feed the robot's sensors back into the simulation and close the loop. A dead brain that can finally feel the world it's walking through. This fly has been dead for years. Its brain runs on my laptop. Its decisions move a machine through my living room. One dead nervous system, walking again.

kiruwaaaa

75,565 次观看 • 8 天前

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo for partnering with me on this post.

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo for partnering with me on this post.

Santiago

73,062 次观看 • 10 个月前

Sharing our computer vision system design for TRON: Ares. We approached these shots from the perspective of the machine. The frame becomes a field of observation. Every pixel is potential information waiting to be interpreted. The process begins by squarepacking the live action plates, dividing the image into a computational lattice. Once the grid exists, the system begins to read. Targets emerge. Motion vectors form. Signals ripple through layers of analysis. Inside these overlays lives the source code of the system itself. Streams of logic, fragments of instruction, the internal language of the machine. At times the imagery dissolves directly into that code, as if the world itself is being rewritten from within. Visual information collapses into structure, then reconstructs again as data. Scanning pulses move across the frame. Timers react. Coordinates lock. Each layer is part of a larger ecosystem of perception, constantly processing, evaluating, and iterating. The goal was maximum detail and living systems. Interfaces that breathe, think, and evolve in real time.

Sharing our computer vision system design for TRON: Ares. We approached these shots from the perspective of the machine. The frame becomes a field of observation. Every pixel is potential information waiting to be interpreted. The process begins by squarepacking the live action plates, dividing the image into a computational lattice. Once the grid exists, the system begins to read. Targets emerge. Motion vectors form. Signals ripple through layers of analysis. Inside these overlays lives the source code of the system itself. Streams of logic, fragments of instruction, the internal language of the machine. At times the imagery dissolves directly into that code, as if the world itself is being rewritten from within. Visual information collapses into structure, then reconstructs again as data. Scanning pulses move across the frame. Timers react. Coordinates lock. Each layer is part of a larger ecosystem of perception, constantly processing, evaluating, and iterating. The goal was maximum detail and living systems. Interfaces that breathe, think, and evolve in real time.

GMUNK

37,783 次观看 • 6 个月前

The Mathematics of Moving a Cursor with Neural Signals What might Neuralink Neuralink be doing Mathematically? Consider the task of moving a cursor without touching it. The machine is not looking for a full thought, a sentence, or an image. For this Control problem, the useful object is an intended movement state. sₜ = (pₜ, vₜ) Here, pₜ is the cursor position at time t, and vₜ is the velocity the user is trying to express. The implant records neural activity through many electrode channels, then the decoder tries to estimate vₜ from that activity. Neuralink’s PRIME material describes the N1 Implant as recording and transmitting brain activity with the goal of enabling computer control. For channel i, a simple population model is rᵢ(t) ≈ bᵢ + aᵢ max(0, dᵢ · vₜ) + ηᵢ(t) where rᵢ(t) is the measured activity, bᵢ is baseline activity, aᵢ is channel gain, dᵢ is the channel’s preferred movement direction, and ηᵢ(t) is noise. One channel is not the command. The useful signal is the pattern across many channels: rₜ = (r₁(t), r₂(t), …, rₙ(t)) The decoder subtracts the baseline vector b and applies a learned map W: v̂ₜ = W(rₜ − b) This gives an estimate of the intended velocity. The cursor then updates by pₜ₊₁ = pₜ + Δt v̂ₜ This is the loop shown in the render: neural activity -> decoded velocity -> cursor motion The cortical network and electrode threads show the measurement side. The N1 Implant is described as using 1,024 electrodes distributed across 64 flexible threads, each thinner than a human hair. The decoder panel shows the computational side with activity rₜ, decoded velocity v̂ₜ, and the cursor state pₜ changing over time. A noisy biological pattern becomes a state estimate. That estimate becomes motion on a screen. Therefore, the first lesson is not that Neuralink makes the brain a screen. For cursor control, the Mathematics is more precise: A small piece of intention is represented as a hidden state, measured through neural activity, decoded as a vector, and turned into action. #Neuralink #BrainComputerInterface #NeuralEngineering #Mathematics #StateEstimation #Neuroscience #MachineLearning #BiomedicalEngineering

The Mathematics of Moving a Cursor with Neural Signals What might Neuralink Neuralink be doing Mathematically? Consider the task of moving a cursor without touching it. The machine is not looking for a full thought, a sentence, or an image. For this Control problem, the useful object is an intended movement state. sₜ = (pₜ, vₜ) Here, pₜ is the cursor position at time t, and vₜ is the velocity the user is trying to express. The implant records neural activity through many electrode channels, then the decoder tries to estimate vₜ from that activity. Neuralink’s PRIME material describes the N1 Implant as recording and transmitting brain activity with the goal of enabling computer control. For channel i, a simple population model is rᵢ(t) ≈ bᵢ + aᵢ max(0, dᵢ · vₜ) + ηᵢ(t) where rᵢ(t) is the measured activity, bᵢ is baseline activity, aᵢ is channel gain, dᵢ is the channel’s preferred movement direction, and ηᵢ(t) is noise. One channel is not the command. The useful signal is the pattern across many channels: rₜ = (r₁(t), r₂(t), …, rₙ(t)) The decoder subtracts the baseline vector b and applies a learned map W: v̂ₜ = W(rₜ − b) This gives an estimate of the intended velocity. The cursor then updates by pₜ₊₁ = pₜ + Δt v̂ₜ This is the loop shown in the render: neural activity -> decoded velocity -> cursor motion The cortical network and electrode threads show the measurement side. The N1 Implant is described as using 1,024 electrodes distributed across 64 flexible threads, each thinner than a human hair. The decoder panel shows the computational side with activity rₜ, decoded velocity v̂ₜ, and the cursor state pₜ changing over time. A noisy biological pattern becomes a state estimate. That estimate becomes motion on a screen. Therefore, the first lesson is not that Neuralink makes the brain a screen. For cursor control, the Mathematics is more precise: A small piece of intention is represented as a hidden state, measured through neural activity, decoded as a vector, and turned into action. #Neuralink #BrainComputerInterface #NeuralEngineering #Mathematics #StateEstimation #Neuroscience #MachineLearning #BiomedicalEngineering

Mathelirium

14,520 次观看 • 4 个月前

this is pure f*cking gold andrej karpathy landed at anthropic 14 weeks ago about that "two anthropic seniors made his loop 1000x better with graph engineering" line - i can't find a source for it anywhere, so i'm not passing it off as fact what is public and checkable: anthropic's own cookbook on building knowledge graphs, plus karpathy's autoresearch loop that ran 700 experiments in 2 days and surfaced 20 optimizations on its own i wired the graph approach into my own setup the difference showed up on the very first reply claude skipped the stock answer and actually reasoned through the problem start with the cookbook, the full write-up is below

this is pure f*cking gold andrej karpathy landed at anthropic 14 weeks ago about that "two anthropic seniors made his loop 1000x better with graph engineering" line - i can't find a source for it anywhere, so i'm not passing it off as fact what is public and checkable: anthropic's own cookbook on building knowledge graphs, plus karpathy's autoresearch loop that ran 700 experiments in 2 days and surfaced 20 optimizations on its own i wired the graph approach into my own setup the difference showed up on the very first reply claude skipped the stock answer and actually reasoned through the problem start with the cookbook, the full write-up is below

Annatar.md

46,855 次观看 • 18 天前

Had my mind blown today by how simple it was to buy Tokenized Gold. I was listening to Balaji talk about Gold today on a podcast with Luke Gromen I asked ChatGPT “what’s the best way to buy Gold that’s digital and almost like owning the real thing?” It responded with “tokenized gold: Each token represents ownership of real, physical gold This is not a gold stock, not an ETF, and not a synthetic bet—it’s closer to digital warehouse receipts for gold.” It then said the best option was Tether Gold (XAUT). I remembered recently the World team saying that you could buy real assets like Gold on World App. I wondered if you could buy XAUT? I opened the app and one of the first ten assets was XAUT. Swap USDC for Tether Gold in under 10 seconds and that’s it. The experience was magical. We are so close the Balaji vision of interoperable money and assets. When everything is a token, value will be the only thing that matters again. Tokens are just 24/7 global markets for trust. People will be able to move in and out of whatever assets they trust instantly. Trust will become the most important thing that compounds. What’s even cooler, is that you can then send Gold to anyone, instantly in World Chat 💬 Any asset, anywhere, to anyone, in any any chat. Check out the screen recordings to see how fast and easy it was

Had my mind blown today by how simple it was to buy Tokenized Gold. I was listening to Balaji talk about Gold today on a podcast with Luke Gromen I asked ChatGPT “what’s the best way to buy Gold that’s digital and almost like owning the real thing?” It responded with “tokenized gold: Each token represents ownership of real, physical gold This is not a gold stock, not an ETF, and not a synthetic bet—it’s closer to digital warehouse receipts for gold.” It then said the best option was Tether Gold (XAUT). I remembered recently the World team saying that you could buy real assets like Gold on World App. I wondered if you could buy XAUT? I opened the app and one of the first ten assets was XAUT. Swap USDC for Tether Gold in under 10 seconds and that’s it. The experience was magical. We are so close the Balaji vision of interoperable money and assets. When everything is a token, value will be the only thing that matters again. Tokens are just 24/7 global markets for trust. People will be able to move in and out of whatever assets they trust instantly. Trust will become the most important thing that compounds. What’s even cooler, is that you can then send Gold to anyone, instantly in World Chat 💬 Any asset, anywhere, to anyone, in any any chat. Check out the screen recordings to see how fast and easy it was

Shane Mac

12,559 次观看 • 8 个月前

i built a tool that lets me never work again in my life it watches smart wallets on robinhood chain and finds the next runner before anyone else knows it exists not joking $100 → $2,640 in 41 minutes STAMPEDE. it watches wallets on robinhood chain the idea is stupid simple: a runner isn't new money. it's a herd of wallets leaving one coin and piling into the next. the chart moves after the herd. STAMPEDE sees the herd every few seconds it: → reads every trade on the robinhood chain launchpads (146,329 in one hour) → catches the move: one wallet sold coin A, then bought coin B, within minutes → counts the herd: 54 wallets making the same move in 8 minutes is a stampede. one wallet buying 9 times is noise → ranks every coin by how many wallets are running into it and how fast that's growing → draws it on a 3D map and flies the camera to the biggest herd → shows the wallets and the transactions under every number so you can check i'm not lying only the launchpad coins on robinhood chain for now, not the whole market runs on a laptop, zero api keys for the replay the video is a recorded replay of one real hour at 20x $100 → $2,640 in 41 minutes. the herd was already in, i was right behind it three of those back to back and $100 is $1,839,000 not joking should i post every stampede it catches here, live?

i built a tool that lets me never work again in my life it watches smart wallets on robinhood chain and finds the next runner before anyone else knows it exists not joking $100 → $2,640 in 41 minutes STAMPEDE. it watches wallets on robinhood chain the idea is stupid simple: a runner isn't new money. it's a herd of wallets leaving one coin and piling into the next. the chart moves after the herd. STAMPEDE sees the herd every few seconds it: → reads every trade on the robinhood chain launchpads (146,329 in one hour) → catches the move: one wallet sold coin A, then bought coin B, within minutes → counts the herd: 54 wallets making the same move in 8 minutes is a stampede. one wallet buying 9 times is noise → ranks every coin by how many wallets are running into it and how fast that's growing → draws it on a 3D map and flies the camera to the biggest herd → shows the wallets and the transactions under every number so you can check i'm not lying only the launchpad coins on robinhood chain for now, not the whole market runs on a laptop, zero api keys for the replay the video is a recorded replay of one real hour at 20x $100 → $2,640 in 41 minutes. the herd was already in, i was right behind it three of those back to back and $100 is $1,839,000 not joking should i post every stampede it catches here, live?

Argona

103,192 次观看 • 12 天前

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵

Jim Fan

165,246 次观看 • 1 年前

This is how you get 15x fable 5 usage. Fable -> Composer 2.5 reading Fable -> GPT 5.5 execution I built CNVS to make agent orchestration dead simple and insanely visual. You can watch fable 5 delegate work to cursor, codex, open code, all your agents. using your exsisting ai subs no api pricing. The canvas is fully voice controlled locally with nvidia parakeet for fast and free, or gpt realtime 2 for a fully conversational jarvis experience. The built in mcp and cli means agentic control is bidirectional they can prompt and spawn each other and read agent states. but there is more I built it from the ground up in swift for native performance on mac os. My old 16gb m1 mbp can easily spin up and control eight agents across multiple canvasses. The hermes integration + remote canvases allow you to run agents fully in the cloud on your VPS think google docs of vibe coding. You can literally turn off your mac and they keep working, open cnvs back up and pick up where you left off. I built a cross agent memory system based on 2026 research so all your agents feel like one brain. its SIMPLE and on demand. This eliminates context bloat, and the research supports a massive uptick in cross agent performance. There is soo much more. PS - I am a father building this project 100% solo in my basement on live stream everyday, my goal is to push vibe coding to the next level PPS - CNVS is a lifetime license becuase you DO NOT need another subscription in your life.

This is how you get 15x fable 5 usage. Fable -> Composer 2.5 reading Fable -> GPT 5.5 execution I built CNVS to make agent orchestration dead simple and insanely visual. You can watch fable 5 delegate work to cursor, codex, open code, all your agents. using your exsisting ai subs no api pricing. The canvas is fully voice controlled locally with nvidia parakeet for fast and free, or gpt realtime 2 for a fully conversational jarvis experience. The built in mcp and cli means agentic control is bidirectional they can prompt and spawn each other and read agent states. but there is more I built it from the ground up in swift for native performance on mac os. My old 16gb m1 mbp can easily spin up and control eight agents across multiple canvasses. The hermes integration + remote canvases allow you to run agents fully in the cloud on your VPS think google docs of vibe coding. You can literally turn off your mac and they keep working, open cnvs back up and pick up where you left off. I built a cross agent memory system based on 2026 research so all your agents feel like one brain. its SIMPLE and on demand. This eliminates context bloat, and the research supports a massive uptick in cross agent performance. There is soo much more. PS - I am a father building this project 100% solo in my basement on live stream everyday, my goal is to push vibe coding to the next level PPS - CNVS is a lifetime license becuase you DO NOT need another subscription in your life.

Max Blade

29,824 次观看 • 2 个月前

Yesterday, I told you about a wall I'd been hitting for years. Today I get to break through it — with two announcements. 👇 𝟭. CreativAI is out of stealth — the SQL layer for Physical and Visual AI; also enabling visual intelligence to be verifiable, reliable, and cost-effective. 𝟮. We're launching a product that works today. Not a waitlist. Not a vision deck. Something you can try right now 👇 Whether you're an individual exploring AI, a developer building the next generation of applications, or an enterprise unlocking the value of visual data, CreativAI is ready for you. Deploy in the cloud, on-premises, or integrate through our APIs—whichever fits your workflow. Grateful to our CCO Waleed, our advisors Rob Ferguson and Abdul Jarrar, and to Google for Startups, AWS Startups, Microsoft for Startups, and NVIDIA for Startups Inception for the support. I've spent my career at the foundations of vision-language AI — research at KAUST, Stanford, Meta FAIR, and Adobe. I contributed to some of the building blocks the field now takes for granted: a linear version of CLIP (ICCV13; and the first vision LLMs — VisualGPT (CVPR22), MiniGPT-4(Arxiv'23, ICLR24). Somewhere along the way, the hardest problem in visual AI moved. Every kind of data got its breakthrough. Documents got search. Tabular data got SQL. Code got GitHub. Each one gave messy data a structure — something you could query, verify, and act on. Visual data never got that SQL like accessibility; the largest data type we produce — over 80% of internet traffic, a billion cameras and climbing. And there's a deeper limit. The models got remarkably good, but the events that matter most in your operation barely exist in pretraining data. A general model may have never seen them — so it can't recognize them in your world. That's what we built at Creativ AI. Point it at anything with a lens — cameras, robots, live streams, or years of archives. Our Data Plating technology turns raw pixels into structure the instant something happens: entities, events, behaviors. Not captions. Not metadata. The video itself, as rows and columns. Live streams as they happen, archives you've had for years —all of it becomes a knowledge base you can query. In plain language for your team. Through APIs for your agents. On-device for your robots, so they can close the loop and act inside your workflows. Every row points back to the moment it came from — so every result is traceable, verifiable, and safe to build on. One structured picture. One source of truth. Everyone reading from the same record. And this is where the pretraining gap closes: you teach it your domain — your events, your entities, the rare cases that matter. CreativAI learns what a general model never could, so the long tail of your world becomes queryable data like everything else

Yesterday, I told you about a wall I'd been hitting for years. Today I get to break through it — with two announcements. 👇 𝟭. CreativAI is out of stealth — the SQL layer for Physical and Visual AI; also enabling visual intelligence to be verifiable, reliable, and cost-effective. 𝟮. We're launching a product that works today. Not a waitlist. Not a vision deck. Something you can try right now 👇 Whether you're an individual exploring AI, a developer building the next generation of applications, or an enterprise unlocking the value of visual data, CreativAI is ready for you. Deploy in the cloud, on-premises, or integrate through our APIs—whichever fits your workflow. Grateful to our CCO Waleed, our advisors Rob Ferguson and Abdul Jarrar, and to Google for Startups, AWS Startups, Microsoft for Startups, and NVIDIA for Startups Inception for the support. I've spent my career at the foundations of vision-language AI — research at KAUST, Stanford, Meta FAIR, and Adobe. I contributed to some of the building blocks the field now takes for granted: a linear version of CLIP (ICCV13; and the first vision LLMs — VisualGPT (CVPR22), MiniGPT-4(Arxiv'23, ICLR24). Somewhere along the way, the hardest problem in visual AI moved. Every kind of data got its breakthrough. Documents got search. Tabular data got SQL. Code got GitHub. Each one gave messy data a structure — something you could query, verify, and act on. Visual data never got that SQL like accessibility; the largest data type we produce — over 80% of internet traffic, a billion cameras and climbing. And there's a deeper limit. The models got remarkably good, but the events that matter most in your operation barely exist in pretraining data. A general model may have never seen them — so it can't recognize them in your world. That's what we built at Creativ AI. Point it at anything with a lens — cameras, robots, live streams, or years of archives. Our Data Plating technology turns raw pixels into structure the instant something happens: entities, events, behaviors. Not captions. Not metadata. The video itself, as rows and columns. Live streams as they happen, archives you've had for years —all of it becomes a knowledge base you can query. In plain language for your team. Through APIs for your agents. On-device for your robots, so they can close the loop and act inside your workflows. Every row points back to the moment it came from — so every result is traceable, verifiable, and safe to build on. One structured picture. One source of truth. Everyone reading from the same record. And this is where the pretraining gap closes: you teach it your domain — your events, your entities, the rare cases that matter. CreativAI learns what a general model never could, so the long tail of your world becomes queryable data like everything else

Mohamed Elhoseiny

39,450 次观看 • 1 个月前

Everyone is sleeping on Meta's SAM 3 release. But it's actually a big deal. Here's why: Companies spend millions paying humans to label images and videos frame by frame. A single autonomous driving dataset? Months of work, hundreds of annotators, millions in cost. Without labeled data, you can't train custom models. Without custom models, you're stuck with generic solutions. This is why most companies never move past pilots. SAM 3 breaks this cycle. First let's look at the evolution: SAM 1 segmented objects when you clicked on them. Revolutionary, but one object at a time. SAM 2 added video tracking with memory. Game-changing, but you still manually prompted every object. SAM 3 changes everything with text prompts. Type "yellow school bus" and it finds ALL of them in your image or video. Not just one. Every instance across thousands of frames. Now here's where people get confused: "Can't I just use GPT-5 or Gemini for this?" No, and here's why that's a terrible approach. Large multimodal LLMs are great for reasoning, but they're slow and expensive for production visual tasks. You're paying API costs per image, waiting seconds for responses, getting inconsistent results. SAM 3 runs in 30 milliseconds on a single GPU for 100+ objects. That's 100x faster, and you own the infrastructure. More importantly, SAM 3 gives you precise pixel-level masks, not descriptions. Try asking an LLM to segment every defective part on a manufacturing line in real-time. It won't work. SAM 3 does this effortlessly. The real breakthrough is their data engine. Meta built an AI-human hybrid system that's 5x faster for complex annotations. They trained SAM 3 on 4 million unique visual concepts - 50x more than existing benchmarks like LVIS. SAM 3 is trained on 4 million unique visual concepts, it handles everything: - Text-based concept search - Interactive refinement with clicks - Video tracking across frames - Zero-shot detection of new concepts The model is open source. Weights, code, and benchmarks are on GitHub. If you're building computer vision applications, this is the foundation model to evaluate. The annotation time savings alone will pay for integration costs within weeks. Find the relevant links in the next tweet!

Everyone is sleeping on Meta's SAM 3 release. But it's actually a big deal. Here's why: Companies spend millions paying humans to label images and videos frame by frame. A single autonomous driving dataset? Months of work, hundreds of annotators, millions in cost. Without labeled data, you can't train custom models. Without custom models, you're stuck with generic solutions. This is why most companies never move past pilots. SAM 3 breaks this cycle. First let's look at the evolution: SAM 1 segmented objects when you clicked on them. Revolutionary, but one object at a time. SAM 2 added video tracking with memory. Game-changing, but you still manually prompted every object. SAM 3 changes everything with text prompts. Type "yellow school bus" and it finds ALL of them in your image or video. Not just one. Every instance across thousands of frames. Now here's where people get confused: "Can't I just use GPT-5 or Gemini for this?" No, and here's why that's a terrible approach. Large multimodal LLMs are great for reasoning, but they're slow and expensive for production visual tasks. You're paying API costs per image, waiting seconds for responses, getting inconsistent results. SAM 3 runs in 30 milliseconds on a single GPU for 100+ objects. That's 100x faster, and you own the infrastructure. More importantly, SAM 3 gives you precise pixel-level masks, not descriptions. Try asking an LLM to segment every defective part on a manufacturing line in real-time. It won't work. SAM 3 does this effortlessly. The real breakthrough is their data engine. Meta built an AI-human hybrid system that's 5x faster for complex annotations. They trained SAM 3 on 4 million unique visual concepts - 50x more than existing benchmarks like LVIS. SAM 3 is trained on 4 million unique visual concepts, it handles everything: - Text-based concept search - Interactive refinement with clicks - Video tracking across frames - Zero-shot detection of new concepts The model is open source. Weights, code, and benchmarks are on GitHub. If you're building computer vision applications, this is the foundation model to evaluate. The annotation time savings alone will pay for integration costs within weeks. Find the relevant links in the next tweet!

Akshay 🚀

46,438 次观看 • 10 个月前