I've been putting together an interactive tool called DiffusionLab... for explaining the geometric intuition behind diffusion and flow based generative models. Sampling is actually being done in the browser using Tensorflow.js! It is still in the very early stages.show more

I made a tool called Diffusion Explorer that lets... you to train and visualize simple 2D diffusion and flow models live in the browser. You can draw your own distributions and observe how the generated samples converge during training. Try it live 👇show more

Still one of my fav projects I worked on... in the day job – an interactive experience showing low-carbon technology for the home Custom 3D models created in Blender 🔶 and everything is rendered in the browser using Three.js and the awesome react-three-fibershow more

MaskINT: Video Editing via Interpolative Non-autoregressive Masked Transformers paper... page: Recent advances in generative AI have significantly enhanced image and video editing, particularly in the context of text prompt control. State-of-the-art approaches predominantly rely on diffusion models to accomplish these tasks. However, the computational demands of diffusion-based methods are substantial, often necessitating large-scale paired datasets for training, and therefore challenging the deployment in practical applications. This study addresses this challenge by breaking down the text-based video editing process into two separate stages. In the first stage, we leverage an existing text-to-image diffusion model to simultaneously edit a few keyframes without additional fine-tuning. In the second stage, we introduce an efficient model called MaskINT, which is built on non-autoregressive masked generative transformers and specializes in frame interpolation between the keyframes, benefiting from structural guidance provided by intermediate frames. Our comprehensive set of experiments illustrates the efficacy and efficiency of MaskINT when compared to other diffusion-based methodologies. This research offers a practical solution for text-based video editing and showcases the potential of non-autoregressive masked generative transformers in this domain.show more

Flow matching produces smooth, deterministic trajectories. In contrast, the... sampling process of a diffusion model is chaotic, resembling the random motion of gas particles.show more

Grok 4.7 works really well with Blender. I’ve been... using them to create these rough Raptor engine 3D models, and the improvement in quality is huge. They’re still early versions, but the details are already coming together beautifully.show more

The latent space of earlier generative models like GANS... can linearly encode concepts of the data. What if the data was model weights? We present weights2weights, a subspace in diffusion weights that behaves as an interpretable latent space over customized diffusion models.show more

There's a lot of both 2D and 3D effort... that we're putting into Star Birds together with our friends at Toukana Interactive! Here's a quick behind the scenes look at some of the assets, 3D models, birb shenanigans and cutscenes that you might have already seen in the Early Access version! 🚀✨ Have you already joined us on this space adventure? Don't miss out! Star Birds in Early Access is available on Steam:show more

This symmetric diffusion paper at ICLR is nice (simple... idea in retrospect): SymmCD: Symmetry-Preserving Crystal Generation with Diffusion Models We'd actually implemented this idea internally at Orbital, and it works nicely even for very large crystal structures:show more

relating to Stocky, this is the game they're being... made for! The game is called Meleeniacs and its in VERY early development. This is all subject to change, but I'm open to any questions about the plans for this game :Dshow more

Keir Starmer explaining how shunting off your kids early... to a Breakfast Club is, wait for it, "bringing families and children together across the country". Clearly he isn't all there in the head.show more

This is a fun interactive article explaining the intuition... behind Gaussian Processes. Gaussian Processes allow you to construct a distribution over functions with a continuous domain. They are incredibly useful for applications like Bayesian regression.show more

We are also releasing self-contained lecture notes that explain... flow matching and diffusion models from scratch. This goes from "zero" to the state-of-the-art in modern Generative AI. 📖 Read the notes here: Joint work with Ezra Erives.show more

Putting something together in NYC on April 21st. World... models, policy learning, robotics - if you're building at the frontier - DM anyone on the General Intuition team to join.show more

Stochastic and deterministic sampling strategies for diffusion models produce... strikingly different trajectories, but both ultimately achieve the same aim. I had a great time presenting our work, Diffusion Explorer, this week at IEEE VIS in Vienna.show more

Kimi K3 built a fully procedural browser-based 3D exploration... game using Three.js WebGPU and GPU compute. It procedurally generated the environment, while using a 3D asset generation tool to create the rider and horse models, producing an expansive open world with forests, a log-cabin village, snowy mountains, and dynamic weather. External assets used: animated cowboy and horse models and terrain data.show more

Canelo 🤝 Usyk Canelo Alvarez and Oleksandr Usyk have... been putting the work in together out in Spain🔥 Who is Usyk training for?show more

Ask Claude to document and describe the main flows... in your app and output in a single page html + json data file. Incredibly useful for humans and the JSON file is very useful for explaining the flow to the LLM when working on new features/bugfixes.show more