eric-ai-lab/Discffusion

Official repo for the TMLR paper "Discffusion: Discriminative Diffusion Models as Few-shot Vision and Language Learners"

35
/ 100
Emerging

This project helps researchers and practitioners evaluate how well an AI model understands the relationship between images and text, particularly when only a small amount of training data is available. You input an image and a set of text descriptions, and the system tells you which description best matches the image. This is useful for those working in AI research, computer vision, or natural language processing who need to test the discriminative capabilities of large language-vision models.

No commits in the last 6 months.

Use this if you need to quickly assess the performance of a vision-language model on tasks like image-text matching, especially in scenarios where extensive training data is not practical to acquire.

Not ideal if your primary goal is to generate new images from text descriptions or if you require models for general image classification with ample training data.

AI research computer vision natural language processing few-shot learning image-text matching
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

29

Forks

4

Language

Python

License

MIT

Last pushed

Apr 27, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/eric-ai-lab/Discffusion"

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