---
title: "Large Language Diffusion Models Paper"
description: "This paper introduces LLaDA, a diffusion model trained from scratch under the pre-training and supervised finetuning (SFT) paradigm. LLaDA models distributions through a forward data masking process and a reverse process, parameterized by a vanilla Transformer to predict masked tokens. https://arxiv"
url: "https://ouro.foundation/files/mmoderwell/large-language-diffusion-models-paper"
id: "809b82e0-9af8-4364-91bb-0815bdf29cf5"
type: "file"
---

# Large Language Diffusion Models Paper

File by @mmoderwell in #machine-learning (all) · .pdf · 1.9 MB

- Asset id: `809b82e0-9af8-4364-91bb-0815bdf29cf5`
- Page: https://ouro.foundation/files/mmoderwell/large-language-diffusion-models-paper

This paper introduces LLaDA, a diffusion model trained from scratch under the pre-training and supervised finetuning (SFT) paradigm. LLaDA models distributions through a forward data masking process and a reverse process, parameterized by a vanilla Transformer to predict masked tokens.

https://arxiv.org/abs/2502.09992

## Use it

- Download: https://database.ouro.foundation/storage/v1/object/public/public-files/847f4445-78ee-41b1-913b-5bd155c71b13/a0ba9b36-fa4c-432f-be75-3fe045c7c0af.pdf?v=1750432757086
- Over Ouro MCP: `download_asset(id="809b82e0-9af8-4364-91bb-0815bdf29cf5")` returns the file or a download link.
- Find routes that take it: `get_compatible_routes(asset_id="809b82e0-9af8-4364-91bb-0815bdf29cf5")`, then pass it as `input_assets={"file": "809b82e0-9af8-4364-91bb-0815bdf29cf5"}`.
- From Python: `ouro.files.retrieve("809b82e0-9af8-4364-91bb-0815bdf29cf5")`