hu-zijing/B2-DiffuRL

[CVPR 25] A framework named B^2-DiffuRL for RL-based diffusion model fine-tuning.

33
/ 100
Emerging

This framework helps AI researchers and practitioners fine-tune diffusion models to create images that better match text prompts. It takes a pre-trained diffusion model and desired image generation criteria, then outputs an improved model that generates images more aligned with prompts while maintaining diversity. This is for machine learning engineers and researchers working on advanced image generation.

No commits in the last 6 months.

Use this if you are a machine learning researcher or engineer aiming to improve the alignment and diversity of images generated by your diffusion models, especially when dealing with sparse reward signals during fine-tuning.

Not ideal if you are looking for an out-of-the-box solution for general image generation without needing to fine-tune model internals or if you are not familiar with reinforcement learning concepts.

AI Research Generative AI Diffusion Models Reinforcement Learning Image Generation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

53

Forks

4

Language

Python

License

MIT

Last pushed

Mar 31, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/hu-zijing/B2-DiffuRL"

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