jaisidhsingh/CoN-CLIP

Implementation of the "Learn No to Say Yes Better" paper.

39
/ 100
Emerging

This project offers an improved way to classify images and understand complex visual relationships by making vision-language models better at handling negative statements. It takes an image and a list of text descriptions (some of which might be negative, like 'this is NOT a cat') and outputs the likelihood that the image matches each description. This is useful for AI researchers and machine learning engineers who are working with advanced image recognition and multimodal understanding tasks.

Use this if you need your image recognition models to more accurately interpret what an image is not, in addition to what it is, improving performance on tasks requiring nuanced understanding.

Not ideal if you are looking for a simple, out-of-the-box image classification tool without needing to integrate it into existing vision-language model pipelines.

AI research image classification natural language understanding computer vision multimodal AI
No Package No Dependents
Maintenance 6 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

40

Forks

4

Language

Python

License

Last pushed

Oct 30, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jaisidhsingh/CoN-CLIP"

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