Ka1b0/Foresight-Guidance

NeurIPS25 Spotlight | Classifier-free guidance (CFG) can be viewed as fixed-point iteration and thus be upgraded.

24
/ 100
Experimental

This project helps researchers and practitioners working with text-to-image diffusion models generate higher-quality images that better match their text descriptions. It takes existing diffusion models and their text prompts as input and produces more visually aligned and efficient image outputs. Anyone involved in developing or fine-tuning advanced image generation systems would find this useful.

Use this if you are a machine learning researcher or engineer looking to improve the image quality and computational efficiency of your text-to-image diffusion models.

Not ideal if you are an end-user simply generating images and not looking to dive into the underlying model guidance mechanisms.

generative-AI image-synthesis diffusion-models AI-research machine-learning-engineering
No Package No Dependents
Maintenance 6 / 25
Adoption 5 / 25
Maturity 13 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

MIT

Last pushed

Nov 24, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Ka1b0/Foresight-Guidance"

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