VectorInstitute/VLDBench

VLDBench: A large-scale benchmark for evaluating Vision-Language Models (VLMs) and Large Language Models (LLMs) on multimodal disinformation detection.

30
/ 100
Emerging

This project helps evaluate how well AI models can detect disinformation in news, using both text and images. It takes news articles paired with images and determines if the content is misleading. Content moderators, policy analysts, and researchers focused on AI safety would use this to understand and improve AI's ability to combat false information.

Use this if you need a comprehensive, human-verified dataset and a standardized method to benchmark AI models for multimodal disinformation detection.

Not ideal if you are looking for an out-of-the-box disinformation detection tool for direct, real-time content moderation.

disinformation-detection content-moderation AI-safety media-analysis regulatory-compliance
No License No Package No Dependents
Maintenance 10 / 25
Adoption 4 / 25
Maturity 8 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

Python

License

Last pushed

Jan 19, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/VectorInstitute/VLDBench"

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