Explainers
No explainer here yet. This is the kind of cell that gets one as soon as something in the queue earns it.
Unread — 4
paper · Jun 2020
Denoising Diffusion Probabilistic Models
The paper that made diffusion work: training collapses to plain noise prediction, and image quality caught up with GANs overnight. Start here.
arXiv 2006.11239 paper · Mar 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamics
Where the idea comes from, borrowed from statistical physics: destroy structure by slow diffusion, then learn the reverse. Five years early, and almost unread.
arXiv 1503.03585 paper · Feb 2025Large Language Diffusion Models (LLaDA)
First 8B masked-diffusion LM matching LLaMA3 8B in-context learning, and beating GPT-4o on reversal tasks.
arXiv 2502.09992 paper · Aug 2025Diffusion Language Models Know the Answer Before Decoding
dLLMs often lock in the right answer halfway through decoding — 3.4× fewer steps, training-free.
arXiv 2508.19982