jinluo12345/Reinforcement-learning-guidance

RLG: Inference-Time Alignment Control for Diffusion Models with Reinforcement Learning Guidance

14
/ 100
Experimental

This project helps image generators fine-tune the alignment of their AI-generated images to specific preferences without needing to retrain their models. It takes an existing diffusion model and an RL-fine-tuned model as input, allowing you to control how closely the output images match desired qualities like aesthetic appeal or text rendering. This tool is ideal for creators, artists, or marketers who use AI to generate images and need precise control over the visual outcomes.

No commits in the last 6 months.

Use this if you generate images with AI diffusion models and want more precise control over how well the generated images align with specific criteria (like text quality or aesthetic scores) without undergoing lengthy retraining processes.

Not ideal if you are looking for a tool to train a new diffusion model from scratch or if you do not work with AI image generation.

AI-image-generation creative-workflows digital-art content-creation image-customization
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 7 / 25
Community 0 / 25

How are scores calculated?

Stars

11

Forks

Language

Python

License

Last pushed

Sep 23, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/jinluo12345/Reinforcement-learning-guidance"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.