Hacker News with Generative AI: Diffusion Models

Why I find diffusion models interesting? (rnikhil.com)
I stumbled across this tweet a week or so back where this company called Inception Labs released a Diffusion LLM (dLLM). Instead of being autoregressive and predicting tokens left to right, here you start all at once and then gradually come up with sensible words simultaneously (start/finish/middle etc. all at once). Something which worked historically for image and video models is now outperforming similar-sized LLMs in code generation.

Generative AI, Machine Learning, Diffusion Models

202 points by whoami_nr 27 days ago | 86 comments

LLM generates the entire output at once (first diffusion LLM) [video] (youtube.com)

Generative AI, Diffusion Models, Artificial Intelligence, Video

4 points by amichail 27 days ago | 0 comments

First large diffusion-based LLM (twitter.com)
Something went wrong, but don’t fret — let’s give it another shot.

Generative AI, Diffusion Models

12 points by mjazwiecki 27 days ago | 1 comments

Large Language Diffusion Models (ml-gsai.github.io)
TL;DR: We introduce LLaDA, a diffusion model with an unprecedented 8B scale, trained entirely from scratch, rivaling LLaMA3 8B in performance.

Diffusion Models, Generative AI

45 points by SerCe 42 days ago | 12 comments

Diffusion Without Tears (notion.site)

Diffusion Models, AI, Generative AI, Computer Vision

62 points by jxmorris12 48 days ago | 20 comments

Stable Flow: Vital Layers for Training-Free Image Editing (omriavrahami.com)
Diffusion models have revolutionized the field of content synthesis and editing. Recent models have replaced the traditional UNet architecture with the Diffusion Transformer (DiT), and employed flow-matching for improved training and sampling. However, they exhibit limited generation diversity. In this work, we leverage this limitation to perform consistent image edits via selective injection of attention features.

Image Editing, Machine Learning, Computer Vision, Artificial Intelligence, Diffusion Models

12 points by gen-ai 65 days ago | 1 comments

Diffusion training from scratch on a micro-budget (github.com/SonyResearch)
This repository provides a minimalistic implementation of our approach to training large-scale diffusion models from scratch on an extremely low budget.

Machine Learning, Diffusion Models, Open Source, Budget Constraints, Research

135 points by lnyan 80 days ago | 23 comments

Pulsar: Secure Steganography for Diffusion Models (iacr.org)
Widespread efforts to subvert access to strong cryptography has renewed interest in steganography, the practice of embedding sensitive messages in mundane cover messages.

Cryptography, Steganography, Security, Artificial Intelligence, Diffusion Models

41 points by aliventer 193 days ago | 3 comments

Diffusion models are real-time game engines (gamengen.github.io)

Diffusion Models, Game Engines, Real-time, Artificial Intelligence

1149 points by jmorgan 218 days ago | 409 comments

DifuzCam: Replacing Camera Lens with a Mask and a Diffusion Model (arxiv.org)

Computer Vision, Machine Learning, Generative AI, Diffusion Models

116 points by dataminer 229 days ago | 47 comments

Diffusion Training from Scratch on a Micro-Budget (arxiv.org)

Diffusion Models, Machine Learning, Computer Vision, Budget

208 points by fzliu 247 days ago | 27 comments

Diffusion Texture Painting (nvidia.com)

Diffusion Models, Generative AI, Image Generation, Computer Graphics, NVIDIA

6 points by jhncls 256 days ago | 0 comments

CAT3D: Create Anything in 3D with Multi-View Diffusion Models (cat3d.github.io)

3D Modeling, AI, Generative AI, Diffusion Models

14 points by alphabetting 321 days ago | 2 comments