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

Diffusion models leverage a variety of samplers. Deterministic methods... like DDIM produce orderly paths. In contrast, stochastic samplers like DDPM produce chaotic trajectories. Despite their differences, both methods draw valid samples from the underlying distribution.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

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models... "Block diffusion sequentially generates blocks of tokens by performing diffusion within each block and conditioning on previous blocks. By combining strength from autoregressive and diffusion models, block diffusion overcomes the limitations of both approaches by supporting variable-length, higher-quality generation and improving inference efficiency with KV caching and parallel sampling."show more

Introducing Diffusion Forcing, which unifies next-token prediction (eg LLMs)... and full-seq. diffusion (eg SORA)! It offers improved performance & new sampling strategies in vision and robotics, such as stable, infinite video generation, better diffusion planning, and more! (1/8)show more

🚨Announcing our #ICLR2025 Oral! 🔥Diffusion LMs are on the... rise for parallel text generation! But unlike autoregressive LMs, they struggle with quality, fixed-length constraints & lack of KV caching. 🚀Introducing Block Diffusion—combining autoregressive and diffusion models for the best of both worlds! 👇1/7show more

Blog post - Autoregression vs Diffusion - Understanding Sampling... via Optimal Transport Autoregression and diffusion look like opposites. Yet, beneath the implementation details, both are solving the same transport problem: how to turn simple noise into structured data.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

This smalldiffusion project is really fun. It is a... tiny library for training and sampling from diffusion models. I love efforts like this, they are much easier to play around with than big libraries like diffusers. Link: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

Diffusion models are sensitive to small changes in the... input noise. We introduce Alias-Free Latent Diffusion Models (𝗔𝗙-𝗟𝗗𝗠) at #CVPR2025. It achieves shift-equivariance and generates consistent outputs. Project: arXiv:show more

1/ Happy to share UniDisc - Unified Multimodal Discrete... Diffusion – We train a 1.5 billion parameter transformer model from scratch on 250 million image/caption pairs using a **discrete diffusion objective**. Our model has all the benefits of diffusion models but now in multimodal space! - flexible compute-quality tradeoff, zero-shot inpainting and editing, better control via classifier-free guidance and lower latency! We open source everything - our code, weights and the training dataset.show more

Excited to share our new work "AvatarPopUp"! "Instant 3D... Human Avatar Generation using Image Diffusion Models" 💻 📄show more

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

Diffusion models have amazing image creation abilities. But how... well does their generative knowledge transfer to discriminative tasks? We present Diffusion Classifier: strong classification results with pretrained conditional diffusion models, *with no additional training*! 1/9show more

⚡ Video diffusion is slow. PDD changes that. 🚀... Introducing Parallel Decoding Distillation, our latest work on accelerating video diffusion models. Paper: Page: By Neta Shaul, Julius Berner, and Chao Liu. Below PDD on LTX-2.3 🔊show more

Selected as a best paper finalist at #CVPR2026: PixelDiT... from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.show more

Introducing Manifold Diffusion Fields (MDF), our new work on... learning generative models over fields defined on curved geometries. This is joint work with our intern Ahmed Elhag (who hasn’t even started his PhD yet!) and @jsusskin at Apple MLR 🧵show more

I didn’t get the chance to greet her, but... I’m feeling both happy and sad at the same time. 🥹❤️ The crazy part is that Shakira stayed at the same hotel where I work and live! So, in a way, I can proudly say that Shakira and I slept under the same roof at least once in our lives.show more