# ADMMDiff

**Repository Path**: koalaaaaaaaaa/ADMMDiff

## Basic Information

- **Project Name**: ADMMDiff
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-11-16
- **Last Updated**: 2025-11-16

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README


# Decoupling Training-Free Guided Diffusion by ADMM
  

## Abstract
In this paper, we consider the conditional generation problem by guiding off-the-shelf unconditional diffusion models with differentiable loss functions in a plug-and-play fashion. We propose a novel framework that distinctly decouples these two components and develop a new algorithm based on the Alternating Direction Method of Multipliers (ADMM) to adaptively balance these components. 

![cover-img](./assets/teaser.png)

## Set environment
```
conda env create -f environment.yml
conda activate admm
```
## Run
Please refer to the folders ./NonLinear, ./Linear and ./Motion.