Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

We've released the code for LegoGPT. This autoregressive model generates physically stable and buildable designs from text prompts, by integrating physics laws and assembly constraints into LLM training and inference. This work is led by PhD students Ava Pun, Kangle Deng, Ruixuan Liu, and in collaboration with CMU faculty... show more

Jun-Yan Zhu

13,681 subscribers

38,730 просмотров • 1 год назад •via X (Twitter)

Искусство Наука и технологии Образование

Anya Rossi• Live Now

Private livecam show

Комментарии: 7

Фото профиля Or Patashnik

Or Patashnik1 год назад

@AvaLovelace0 @kangle_deng Wow, really cool!

Фото профиля Rainmaker

Rainmaker2 лет назад

Here I share an XGBoost model that delivers a 25% CAGR with minimal drawdown on Visa stock. In this free Substack post I share code and commentary for a powerful Machine Learning strategy that delivers powerful returns.

Фото профиля Jason Liu

Jason Liu1 год назад

@AvaLovelace0 @kangle_deng Awesome project 👍🏼. Some designs may require other orientation than from the ground up. I’m excited to learn about this!

Фото профиля Redcrown

Redcrown1 год назад

@AvaLovelace0 @kangle_deng woah, this is soo cool

Фото профиля Ant A

Ant A1 год назад

@AvaLovelace0 @kangle_deng So just GenAI every step/layer?

Фото профиля Aiden

Aiden1 год назад

@AvaLovelace0 @kangle_deng Super interesting project! We're also big believers in using natural language to create. With jenova ai, anyone can build their own custom AI apps just by describing what they need.

Фото профиля Max Zhaoshuo Li 李赵硕

Max Zhaoshuo Li 李赵硕1 год назад

@AvaLovelace0 @kangle_deng Very interesting work! Congrats!

Похожие видео

LegoGPT, an LLM-based system that generates physically stable LEGO structures from text prompts, backed by a new 47,000+ sample dataset and physics-aware filtering during inference. → LegoGPT is trained on a custom dataset, StableText2Lego, which includes 47,000+ 3D LEGO models mapped to text, spanning 28,000+ unique objects. → The model predicts LEGO bricks sequentially like tokens, using next-token prediction in a transformer setup. → To ensure physical stability, LegoGPT integrates physics-aware rollback and validity filtering, pruning out structurally invalid brick placements. → The generated designs are aesthetically aligned with prompts, physically buildable, and tested both with human manual assembly and robotic arms. → The team also introduced a text-driven LEGO coloring/texturing pipeline, enabling more expressive and customized outputs. → The dataset, code, and models are all publicly released under an open-access license.

LegoGPT, an LLM-based system that generates physically stable LEGO structures from text prompts, backed by a new 47,000+ sample dataset and physics-aware filtering during inference. → LegoGPT is trained on a custom dataset, StableText2Lego, which includes 47,000+ 3D LEGO models mapped to text, spanning 28,000+ unique objects. → The model predicts LEGO bricks sequentially like tokens, using next-token prediction in a transformer setup. → To ensure physical stability, LegoGPT integrates physics-aware rollback and validity filtering, pruning out structurally invalid brick placements. → The generated designs are aesthetically aligned with prompts, physically buildable, and tested both with human manual assembly and robotic arms. → The team also introduced a text-driven LEGO coloring/texturing pipeline, enabling more expressive and customized outputs. → The dataset, code, and models are all publicly released under an open-access license.

Rohan Paul

75,281 просмотров • 1 год назад

🔥Excited to share the first released work from our IEI lab! Congrats to AnteaWu 🎉 This work is motivated by the lack of quantitative evaluation for physics alignment in video world models. With tools like MegaSam and CoTracker, we can directly reconstruct dynamic 3D scenes, enabling quantitative evaluation of physical alignment. Both code and data are released — feel free to try it out! It should work, but if it doesn’t, contact AnteaWu directly : )

🔥Excited to share the first released work from our IEI lab! Congrats to AnteaWu 🎉 This work is motivated by the lack of quantitative evaluation for physics alignment in video world models. With tools like MegaSam and CoTracker, we can directly reconstruct dynamic 3D scenes, enabling quantitative evaluation of physical alignment. Both code and data are released — feel free to try it out! It should work, but if it doesn’t, contact AnteaWu directly : )

Xueyan Zou

10,217 просмотров • 3 месяцев назад

[1/2] We’ve released the code for #pix2pixturbo and #CycleGANTurbo. These conditional GANs are able to adapt a text-to-image model such as SD-Turbo for both paired and unpaired image translation with a single step (0.11 sec on A100 and 0.29 sec on A6000). Try our code and the Gradio demo. Paper: Code: Demo: This is a joint work with Gaurav Parmar (the leading author), Taesung Park, and Srinivasa Narasimhan. This work shows that a pre-trained one-step model can be easily adapted to conditional GANs frameworks for downstream image editing and synthesis tasks. #Edges2Cats

[1/2] We’ve released the code for #pix2pixturbo and #CycleGANTurbo. These conditional GANs are able to adapt a text-to-image model such as SD-Turbo for both paired and unpaired image translation with a single step (0.11 sec on A100 and 0.29 sec on A6000). Try our code and the Gradio demo. Paper: Code: Demo: This is a joint work with Gaurav Parmar (the leading author), Taesung Park, and Srinivasa Narasimhan. This work shows that a pre-trained one-step model can be easily adapted to conditional GANs frameworks for downstream image editing and synthesis tasks. #Edges2Cats

Jun-Yan Zhu

36,509 просмотров • 2 лет назад

Today, we are releasing Stable Video Diffusion, our first foundation model for generative AI video based on the image model, Stable Diffusion. As part of this research preview, the code, weights, and research paper are now available. Additionally, today you can sign up for our waitlist to access a new upcoming web experience featuring a Text-To-Video interface. To access the model & sign up for our waitlist, visit our website here:

Today, we are releasing Stable Video Diffusion, our first foundation model for generative AI video based on the image model, Stable Diffusion. As part of this research preview, the code, weights, and research paper are now available. Additionally, today you can sign up for our waitlist to access a new upcoming web experience featuring a Text-To-Video interface. To access the model & sign up for our waitlist, visit our website here:

Stability AI

1,024,725 просмотров • 2 лет назад

Wow! AI ASSISTED GARAGE MANUFACTURING IS ABOUT TO EXPLODE! CAD Drawings From Just A Picture! MIT just released something profound for creators and engineers alike. Picture this. You take a photo of an object, upload it, and an AI delivers a fully parametric CAD model, complete with editable code and construction history. This is open source GenCAD, from MIT's Decode Lab. It uses autoregressive transformers and diffusion models, trained on hundreds of thousands of images and CAD files. Input a 2D photo or sketch. Output valid CadQuery Python code that beats models like GPT-4.5 in accuracy. Why does this matter? It speeds up reverse engineering, prototyping, and part searches in vast databases. No more hours spent modeling from scratch. Field repairs, custom designs, education, all transformed. It even retrieves similar parts from libraries of thousands. For industries like manufacturing and aerospace, it cuts costs and boosts innovation. Hobbyists gain pro tools without the steep curve. I am testing it now on random objects and can not believe how much of a super power this is. I can start dozens of companies just on this AI model. This open-source gem is here: The future of building stuff arrives in a snapshot.

Wow! AI ASSISTED GARAGE MANUFACTURING IS ABOUT TO EXPLODE! CAD Drawings From Just A Picture! MIT just released something profound for creators and engineers alike. Picture this. You take a photo of an object, upload it, and an AI delivers a fully parametric CAD model, complete with editable code and construction history. This is open source GenCAD, from MIT's Decode Lab. It uses autoregressive transformers and diffusion models, trained on hundreds of thousands of images and CAD files. Input a 2D photo or sketch. Output valid CadQuery Python code that beats models like GPT-4.5 in accuracy. Why does this matter? It speeds up reverse engineering, prototyping, and part searches in vast databases. No more hours spent modeling from scratch. Field repairs, custom designs, education, all transformed. It even retrieves similar parts from libraries of thousands. For industries like manufacturing and aerospace, it cuts costs and boosts innovation. Hobbyists gain pro tools without the steep curve. I am testing it now on random objects and can not believe how much of a super power this is. I can start dozens of companies just on this AI model. This open-source gem is here: The future of building stuff arrives in a snapshot.

Brian Roemmele

121,948 просмотров • 6 месяцев назад

For science, AI sovereignty and physics-grounded reasoning are non-negotiable. But how can we teach a small LLM like Gemma-4-E4B physics? One way is to use Agent Skills, but this has so far been limited to closed frontier models. mistral․rs now implements Agent Skills natively: the first self-hosted inference engine that does this as part of the local inference substrate, where we can use small models to solve complex scientific and other tasks in a flexible and scalable way. We are in a period of uncertainty about frontier models - access, pricing, deprecation, abrupt restriction. The good news is that when the entire stack runs locally we can build AI that is entirely your own: You own the weights, the skills, the execution loop, the data - all of it runs on your hardware and is reproducible and durable. While virtually all local inference engines expose a model behind an OpenAI-compatible endpoint, everything agentic is then assembled around it by an external orchestrator that injects context, manages tools, mounts files, and brokers execution. mistral․rs is natively agentic and moves that machinery into the server itself, allowing us to build complex agentic workflows and run them locally, on open-source models. With this new feature you can now upload Agent Skills bundles to /v1/skills, reference them from Responses API requests by identity, and run them inside a native agentic loop with persistent Python sessions, figure capture, sandboxed shell execution, file inputs mounted directly into the working session; plug-and-play and completely compatible with your existing code/workflow. A model with a native skill substrate can act, observe consequences, and can modify what it is able to do. The skill is retained procedural capability of the system. Attached is a short video of all of it: skills, code execution, the full agentic loop carried by Gemma-4-E4B; running entirely on my MacBook Pro. You can install and run a server with this capability in two lines in your terminal, with any quantization you need. Nice work by the Google Gemma team Logan Kilpatrick Demis Hassabis and Eric Buehler with mistral․rs!

For science, AI sovereignty and physics-grounded reasoning are non-negotiable. But how can we teach a small LLM like Gemma-4-E4B physics? One way is to use Agent Skills, but this has so far been limited to closed frontier models. mistral․rs now implements Agent Skills natively: the first self-hosted inference engine that does this as part of the local inference substrate, where we can use small models to solve complex scientific and other tasks in a flexible and scalable way. We are in a period of uncertainty about frontier models - access, pricing, deprecation, abrupt restriction. The good news is that when the entire stack runs locally we can build AI that is entirely your own: You own the weights, the skills, the execution loop, the data - all of it runs on your hardware and is reproducible and durable. While virtually all local inference engines expose a model behind an OpenAI-compatible endpoint, everything agentic is then assembled around it by an external orchestrator that injects context, manages tools, mounts files, and brokers execution. mistral․rs is natively agentic and moves that machinery into the server itself, allowing us to build complex agentic workflows and run them locally, on open-source models. With this new feature you can now upload Agent Skills bundles to /v1/skills, reference them from Responses API requests by identity, and run them inside a native agentic loop with persistent Python sessions, figure capture, sandboxed shell execution, file inputs mounted directly into the working session; plug-and-play and completely compatible with your existing code/workflow. A model with a native skill substrate can act, observe consequences, and can modify what it is able to do. The skill is retained procedural capability of the system. Attached is a short video of all of it: skills, code execution, the full agentic loop carried by Gemma-4-E4B; running entirely on my MacBook Pro. You can install and run a server with this capability in two lines in your terminal, with any quantization you need. Nice work by the Google Gemma team Logan Kilpatrick Demis Hassabis and Eric Buehler with mistral․rs!

Markus J. Buehler

10,229 просмотров • 2 месяцев назад

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:

Max Zhaoshuo Li 李赵硕

1,078,929 просмотров • 3 месяцев назад

We've been ramping up usage of AI tools on our design team at Coinbase 🛡️ Two examples: 1. Write a text prompt and get a figma mockup as a starting point for your design 2. Click a button and turn any figma design into front-end code The front-end code it generates adheres to our design system, and is trained on our library of UIs Shout out to 🛡️ tali krakowsky apel🛡️ Blair McKee and the team pushing this forward.

We've been ramping up usage of AI tools on our design team at Coinbase 🛡️ Two examples: 1. Write a text prompt and get a figma mockup as a starting point for your design 2. Click a button and turn any figma design into front-end code The front-end code it generates adheres to our design system, and is trained on our library of UIs Shout out to 🛡️ tali krakowsky apel🛡️ Blair McKee and the team pushing this forward.

Brian Armstrong

520,460 просмотров • 2 лет назад

An interactive world model developed by NVIDIA in collaboration with academic partners. - DreamDojo turns egocentric human video data into physical intelligence. - Human data is more scalable than robotics data but lacks action labels. - To solve this, a dedicated action model extracts latent actions by identifying physics and motion deltas between frames. Training - A massive 44k hours of video data are used for pre-training. - Post-training on small-scale robot datasets maps human physics to specific robot embodiments. - An additional distillation stage converts the model into an autoregressive, few-step diffusion model, enabling real-time, action-controllable simulation. Primary Use Cases - Live Teleoperation: Controlling a robot inside a world simulation in real-time. - Model-based Planning: Previewing and curating the best actions for improved success. - Policy Evaluation: Testing robot policies in realistic, out-of-distribution scenarios. Everything that's open-sourced: weights, code, post-training dataset, eval set, and details to reproduce.

An interactive world model developed by NVIDIA in collaboration with academic partners. - DreamDojo turns egocentric human video data into physical intelligence. - Human data is more scalable than robotics data but lacks action labels. - To solve this, a dedicated action model extracts latent actions by identifying physics and motion deltas between frames. Training - A massive 44k hours of video data are used for pre-training. - Post-training on small-scale robot datasets maps human physics to specific robot embodiments. - An additional distillation stage converts the model into an autoregressive, few-step diffusion model, enabling real-time, action-controllable simulation. Primary Use Cases - Live Teleoperation: Controlling a robot inside a world simulation in real-time. - Model-based Planning: Previewing and curating the best actions for improved success. - Policy Evaluation: Testing robot policies in realistic, out-of-distribution scenarios. Everything that's open-sourced: weights, code, post-training dataset, eval set, and details to reproduce.

The Humanoid Hub

11,575 просмотров • 6 месяцев назад

NVIDIA just released a new open source transcription model, Nemotron Speech ASR, designed from the ground up for low-latency use cases like voice agents. Here's a voice agent built with this new model. 24ms transcription finalization and total voice-to-voice inference time under 500ms. This agent actually uses *three* NVIDIA open source models: - Nemotron Speech ASR - Nemotron 3 Nano 30GB in a 4-bit quant (released in December) - A preview checkpoint of the upcoming Magpie text-to-speech model These models are all truly open source: weights, training data, training code, and inference code. This is a big deal! Jensen said in the CES keynote yesterday that he expects open source models to catch up to proprietary models this year in a number of categories. NVIDIA is putting their weight behind making this happen. (As Alan Kay said, the best way to predict the future is to invent it.) The code for this agent is open source too, of course. You can deploy it to production with Modal and Pipecat AI cloud, or run locally on an NVIDIA DGX Spark or RTX 5090.

NVIDIA just released a new open source transcription model, Nemotron Speech ASR, designed from the ground up for low-latency use cases like voice agents. Here's a voice agent built with this new model. 24ms transcription finalization and total voice-to-voice inference time under 500ms. This agent actually uses three NVIDIA open source models: - Nemotron Speech ASR - Nemotron 3 Nano 30GB in a 4-bit quant (released in December) - A preview checkpoint of the upcoming Magpie text-to-speech model These models are all truly open source: weights, training data, training code, and inference code. This is a big deal! Jensen said in the CES keynote yesterday that he expects open source models to catch up to proprietary models this year in a number of categories. NVIDIA is putting their weight behind making this happen. (As Alan Kay said, the best way to predict the future is to invent it.) The code for this agent is open source too, of course. You can deploy it to production with Modal and Pipecat AI cloud, or run locally on an NVIDIA DGX Spark or RTX 5090.

kwindla

274,679 просмотров • 8 месяцев назад

Can we compile matter - for instance, a pine cone - and derive new active materials, end-to-end from observation to manufacturing? If physical systems can be formalized as composable mathematics, we can point AI that has been shown to resolve long-open mathematical problems at matter itself. Our new work turns bioinspired engineering from analogy into formal compilation: biology and mechanics become explicit, checkable, and executable, so AI reasoning can produce physical designs. This is the first end-to-end demonstration in which a formally compositional multiscale model is carried from a biological hierarchy, through engineered design and fabrication specification, to executable manufacturing code - and then to a physically tested artifact. Background: Humans have long been inspired by biology to advance technology, but this has usually been an ad hoc process rather than a mathematically rigorous one. Natural materials such as pinecones achieve adaptive behavior through mechanisms organized across many scales. Engineering typically translates those mechanisms by analogy: identify a biological principle, build something inspired by it, and validate each new design as a separate case. This can produce remarkable results, but the knowledge does not readily compound. Instead, we represent each scale as a dynamical module with explicit states, stimuli, governing laws, and interfaces. Every scale-to-scale map must preserve the stimulus - response dynamics: evolve the fine-scale system and then map upward, or map upward first and then evolve. The two paths must agree. Because this condition is preserved under composition, locally valid interfaces remain consistent when assembled into the full hierarchy. We then carry that structure into an engineered system, translate the target behavior into a verified fabrication specification, and compile it into G-code: the toolpaths, deposition sequence, temperatures, speeds, and other commands executed by a 3D printer. The intermediate translations are explicit, checkable, and executable rather than completed through an ad hoc handoff. The formal guarantee is that given valid local models and interfaces, their composition remains valid. Whether those models and manufacturing assumptions accurately capture physical reality remains an empirical question. That is why we fabricated and tested the results. We generated four actuator classes by crossing two stimuli - humidity and heat - with two responses: bending and twisting. The fourth, thermal twisting, required no new pipeline and no separate derivation within the framework. It emerged by composing a thermal stimulus module already validated in one case with a twisting module validated in another. The generated G-code produced the intended motion without manual redesign, and all four predictions fell within one experimental standard deviation of the measured response. Why this matters: 1⃣For AI in science, this provides a physics-aware type system against which generative proposals can be checked - and rejected at the interface - before expensive simulation, fabrication, or experiment. It is roughly analogous to proof checking, but for the composition of physical mechanisms. 2⃣For engineering, the accessible design space can scale with a library of validated components rather than with the number of individually derived cases. 3⃣The mathematics, category theory, carries all the way into a physical object on a print bed. This points toward scientific knowledge as executable infrastructure: models that are not only described in papers, but typed, composable, verifiable, and able to compile into experiments. Excellent work led by my student Lee Marom with Skylar Tibbits & Gioele Zardini. Paper published in J. Mech. Phys. Solids along with code, Grasshopper scripts, and manufacturing G-code below.

Can we compile matter - for instance, a pine cone - and derive new active materials, end-to-end from observation to manufacturing? If physical systems can be formalized as composable mathematics, we can point AI that has been shown to resolve long-open mathematical problems at matter itself. Our new work turns bioinspired engineering from analogy into formal compilation: biology and mechanics become explicit, checkable, and executable, so AI reasoning can produce physical designs. This is the first end-to-end demonstration in which a formally compositional multiscale model is carried from a biological hierarchy, through engineered design and fabrication specification, to executable manufacturing code - and then to a physically tested artifact. Background: Humans have long been inspired by biology to advance technology, but this has usually been an ad hoc process rather than a mathematically rigorous one. Natural materials such as pinecones achieve adaptive behavior through mechanisms organized across many scales. Engineering typically translates those mechanisms by analogy: identify a biological principle, build something inspired by it, and validate each new design as a separate case. This can produce remarkable results, but the knowledge does not readily compound. Instead, we represent each scale as a dynamical module with explicit states, stimuli, governing laws, and interfaces. Every scale-to-scale map must preserve the stimulus - response dynamics: evolve the fine-scale system and then map upward, or map upward first and then evolve. The two paths must agree. Because this condition is preserved under composition, locally valid interfaces remain consistent when assembled into the full hierarchy. We then carry that structure into an engineered system, translate the target behavior into a verified fabrication specification, and compile it into G-code: the toolpaths, deposition sequence, temperatures, speeds, and other commands executed by a 3D printer. The intermediate translations are explicit, checkable, and executable rather than completed through an ad hoc handoff. The formal guarantee is that given valid local models and interfaces, their composition remains valid. Whether those models and manufacturing assumptions accurately capture physical reality remains an empirical question. That is why we fabricated and tested the results. We generated four actuator classes by crossing two stimuli - humidity and heat - with two responses: bending and twisting. The fourth, thermal twisting, required no new pipeline and no separate derivation within the framework. It emerged by composing a thermal stimulus module already validated in one case with a twisting module validated in another. The generated G-code produced the intended motion without manual redesign, and all four predictions fell within one experimental standard deviation of the measured response. Why this matters: 1⃣For AI in science, this provides a physics-aware type system against which generative proposals can be checked - and rejected at the interface - before expensive simulation, fabrication, or experiment. It is roughly analogous to proof checking, but for the composition of physical mechanisms. 2⃣For engineering, the accessible design space can scale with a library of validated components rather than with the number of individually derived cases. 3⃣The mathematics, category theory, carries all the way into a physical object on a print bed. This points toward scientific knowledge as executable infrastructure: models that are not only described in papers, but typed, composable, verifiable, and able to compile into experiments. Excellent work led by my student Lee Marom with Skylar Tibbits & Gioele Zardini. Paper published in J. Mech. Phys. Solids along with code, Grasshopper scripts, and manufacturing G-code below.

Markus J. Buehler

128,390 просмотров • 1 месяц назад

AN OXFORD STUDENT IS RUNNING A PARTICLE SIMULATION WITH REAL PEOPLE'S NAMES AND CLAIMS CERN IS TAUNTING HIM THROUGH THE CODE Thousands of particles on a black screen - each one labeled with a real person's name - moving according to the laws of physics in real time and he is completely convinced this is not a simulation but a personal message from CERN directed at him specifically. Particle simulation with collision detection, velocity vectors and brownian motion - technically flawless code that tracks every particle individually and renders trajectories at 60 fps. CERN operates a 17km collider that accelerates protons to 99.9999991% the speed of light and generates a petabyte of data every single day - and apparently found the time to encode Oxford student names into a simulation. The code is real. The physics is correct. The conclusions are a separate conversation.

AN OXFORD STUDENT IS RUNNING A PARTICLE SIMULATION WITH REAL PEOPLE'S NAMES AND CLAIMS CERN IS TAUNTING HIM THROUGH THE CODE Thousands of particles on a black screen - each one labeled with a real person's name - moving according to the laws of physics in real time and he is completely convinced this is not a simulation but a personal message from CERN directed at him specifically. Particle simulation with collision detection, velocity vectors and brownian motion - technically flawless code that tracks every particle individually and renders trajectories at 60 fps. CERN operates a 17km collider that accelerates protons to 99.9999991% the speed of light and generates a petabyte of data every single day - and apparently found the time to encode Oxford student names into a simulation. The code is real. The physics is correct. The conclusions are a separate conversation.

Noisy

5,724,231 просмотров • 3 месяцев назад

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

AI at Meta

150,455 просмотров • 1 год назад

Super clean and efficient meshes by an AI? YES! The typical 3D Generative AI solutions produce lots of artifacts and usually way to many polygons due to volumetric approaches. In comparison “MeshGPT creates triangle meshes by autoregressively sampling from a transformer model that has been trained to produce tokens from a learned geometric vocabulary. These tokens can then be decoded into the faces of a triangle mesh. This method generates clean, coherent, and compact meshes, characterized by sharp edges and high fidelity.” Surely it is limited by the trained vocabulary but various versions can be trained for specific sets to create generative model libraries for certain object groups. Very promising approach with the high quality.

Super clean and efficient meshes by an AI? YES! The typical 3D Generative AI solutions produce lots of artifacts and usually way to many polygons due to volumetric approaches. In comparison “MeshGPT creates triangle meshes by autoregressively sampling from a transformer model that has been trained to produce tokens from a learned geometric vocabulary. These tokens can then be decoded into the faces of a triangle mesh. This method generates clean, coherent, and compact meshes, characterized by sharp edges and high fidelity.” Surely it is limited by the trained vocabulary but various versions can be trained for specific sets to create generative model libraries for certain object groups. Very promising approach with the high quality.

René Schulte

20,772 просмотров • 2 лет назад

You say it. A robot builds it. 🗣️ MIT researchers just showed a system where spoken language turns directly into physical objects. Say “I want a simple stool,” and a robotic arm designs and assembles it in minutes. 🪑 The pipeline is wild but elegant: speech → language model → 3D generative design → voxelized structure → robotic assembly. No CAD, no programming, no manufacturing expertise needed. Unlike 3D printing, this is fast, modular, and reversible. Objects are built from reusable parts, meaning they can be disassembled and turned into something else later. This feels like an early glimpse of “physical AI” becoming an interface: humans describe intent, machines handle geometry, planning, and fabrication. Read more about the paper here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

You say it. A robot builds it. 🗣️ MIT researchers just showed a system where spoken language turns directly into physical objects. Say “I want a simple stool,” and a robotic arm designs and assembles it in minutes. 🪑 The pipeline is wild but elegant: speech → language model → 3D generative design → voxelized structure → robotic assembly. No CAD, no programming, no manufacturing expertise needed. Unlike 3D printing, this is fast, modular, and reversible. Objects are built from reusable parts, meaning they can be disassembled and turned into something else later. This feels like an early glimpse of “physical AI” becoming an interface: humans describe intent, machines handle geometry, planning, and fabrication. Read more about the paper here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

33,069 просмотров • 9 месяцев назад

You might have seen the WuBOT performing at the 2026 Spring Festival Gala; however, most high-dynamic extreme motions you see are executed by overfitted tracking policies. Until now, training a unified policy capable of performing various extreme motions with a high success rate remained an unsolved challenge. We spent an entire year digging into the barrier between general tracking and extreme physical behaviors. After burning through dozens of G1 robots, we finally identified the bottleneck of learning and physical executability. With these discoveries, we developed OmniXtreme: the first general policy that can execute diverse extreme motions, including consecutive flips, extreme balancing, and even breakdancing with rapid contact switches! This capability is achieved by pre-training a flow-based generative control policy and then post-training with actuation-aware residual RL for complex physical dynamics—a step we found critical for successful real-world transfer. This work is a joint collaboration with Unitree. Together, we are pushing the physical limits of humanoid robots. It is incredibly exciting to see a general "robot gymnast" and "robot breakdancer" come to life! It was also our first time publishing a paper with XingXing, which was an enlightening experience. The model checkpoints are now released—we welcome you to play with them! 📦 📄 Paper: 🌐 Project: 💻 Code:

You might have seen the WuBOT performing at the 2026 Spring Festival Gala; however, most high-dynamic extreme motions you see are executed by overfitted tracking policies. Until now, training a unified policy capable of performing various extreme motions with a high success rate remained an unsolved challenge. We spent an entire year digging into the barrier between general tracking and extreme physical behaviors. After burning through dozens of G1 robots, we finally identified the bottleneck of learning and physical executability. With these discoveries, we developed OmniXtreme: the first general policy that can execute diverse extreme motions, including consecutive flips, extreme balancing, and even breakdancing with rapid contact switches! This capability is achieved by pre-training a flow-based generative control policy and then post-training with actuation-aware residual RL for complex physical dynamics—a step we found critical for successful real-world transfer. This work is a joint collaboration with Unitree. Together, we are pushing the physical limits of humanoid robots. It is incredibly exciting to see a general "robot gymnast" and "robot breakdancer" come to life! It was also our first time publishing a paper with XingXing, which was an enlightening experience. The model checkpoints are now released—we welcome you to play with them! 📦 📄 Paper: 🌐 Project: 💻 Code:

Siyuan Huang

107,821 просмотров • 6 месяцев назад

Building and prototyping games is easier than ever with Sidekick. Our designer built a 3D version of this OpenAI GPT-o1 game demo in <1 hour. Used Autodesk Maya for assets, Unity for engine, @bezi_3d Sidekick for scripting & implementation. Sidekick analyzes your project for real-time context of the asset folder, scene objects, components, and more. This allows it to implement simple mechanics—like movement controls and basic physics—for your characters and objects in seconds. It's the AI tool built by game developers, for game developers.

Building and prototyping games is easier than ever with Sidekick. Our designer built a 3D version of this OpenAI GPT-o1 game demo in <1 hour. Used Autodesk Maya for assets, Unity for engine, @bezi_3d Sidekick for scripting & implementation. Sidekick analyzes your project for real-time context of the asset folder, scene objects, components, and more. This allows it to implement simple mechanics—like movement controls and basic physics—for your characters and objects in seconds. It's the AI tool built by game developers, for game developers.

Bezi

15,295 просмотров • 1 год назад

I am testing the amazing Enoch by Mike HealthRanger and team and it is absolutely amazing. A free local AI built on a far wider field of wisdom. My goal: get it to run offline on an iPhone! I am quantizing the model to an aggressive 2-bit level using Q2_K or Q4_K_M. We lose some resolution but we gain portability to any device. The full free version is still small but requires more hefty hardware. But look at this emergency nuclear Fallout response. The keeper-of-the-status-quo Wikipedia types would have you suffer with limited crude ways to react. This response is stellar and will save your life. I will be working on the quantized version and if it is worthy release it for free on the MIT license like the larger Enoch. I will also incorporate this model with all of my other free models as it is a goldmine of astonishing work. Thank you Mike amazing work sir.

I am testing the amazing Enoch by Mike HealthRanger and team and it is absolutely amazing. A free local AI built on a far wider field of wisdom. My goal: get it to run offline on an iPhone! I am quantizing the model to an aggressive 2-bit level using Q2_K or Q4_K_M. We lose some resolution but we gain portability to any device. The full free version is still small but requires more hefty hardware. But look at this emergency nuclear Fallout response. The keeper-of-the-status-quo Wikipedia types would have you suffer with limited crude ways to react. This response is stellar and will save your life. I will be working on the quantized version and if it is worthy release it for free on the MIT license like the larger Enoch. I will also incorporate this model with all of my other free models as it is a goldmine of astonishing work. Thank you Mike amazing work sir.

Brian Roemmele

66,592 просмотров • 11 месяцев назад