Y Hacker News new | ask | show | jobs

user: nathan-barry created: 2023-05-08 karma: 547 https://nathan.rs submissions: Language Modeling Without Neural Networks 2 points | 0 comments Quines in Every Programming Language (Rosetta Code) 2 points | 0 comments Mercury Edit 2: Fastest next-edit prediction with a diffusion LLM (221ms) 1 points | 0 comments Noise schedules considered harmful (2024) 1 points | 0 comments Show HN: Diffuji – a diffusion-powered instant camera 8 points | 5 comments Generating Shakespeare Without Neural Networks 5 points | 0 comments Recursion and Induction: A Self-Contained Course Using ACL2 2 points | 0 comments Fidenae amphitheatre collapse (27 AD) – 20k dead 2 points | 0 comments Show HN: Tiny Diffusion – Minimal diffusion LM in 364 lines 2 points | 0 comments Anchored Diffusion Language Model 2 points | 0 comments Text Diffusion Models Are Faster at Writing Code 3 points | 0 comments Inside ThunderKittens' Python Bindings 2 points | 0 comments Kickstart.nvim: A minimal, single-file starting point for Neovim configuration 1 points | 0 comments How we made AWS Trainium 17x faster (for conv1d) 1 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Show HN: Tiny Diffusion – A character-level text diffusion model from scratch 172 points | 20 comments The Manuscripts of Edsger W. Dijkstra 268 points | 125 comments Streaming DiLoCo: Towards a Distributed Free Lunch 2 points | 0 comments Fast-DLLM: Training-Free Acceleration of Diffusion LLM 70 points | 4 comments BERT is just a single text diffusion step 455 points | 110 comments Rustree: Visualize Rust ASTs in the Browser 6 points | 1 comments Diffusion is not necessarily Spectral Autoregression 3 points | 0 comments 0 points | 0 comments The End of Computing Science (2000) – Edsger W. Dijkstra [pdf] 1 points | 1 comments Planner: Generating Diversified Paragraph via Latent Language Diffusion Model 1 points | 0 comments 0 points | 0 comments