AIFEG/BenchLMM

[ECCV 2024] BenchLMM: Benchmarking Cross-style Visual Capability of Large Multimodal Models

35
/ 100
Emerging

This project helps researchers and developers evaluate how well large multimodal models (LMMs) understand and answer questions about images that vary widely in style, like medical scans, cartoons, or infrared photos. You provide the LMM's text responses to visual questions across these diverse image styles, and the project outputs a performance score indicating the model's 'cross-style' visual understanding capability. This tool is for AI researchers and machine learning engineers developing or comparing LMMs.

No commits in the last 6 months.

Use this if you need to rigorously benchmark and compare the visual comprehension of different large multimodal models across a wide range of image styles and domains.

Not ideal if you are looking for a tool to train or fine-tune an LMM, or if you only work with a single, consistent image style.

AI model evaluation Multimodal AI Computer vision benchmarking Visual question answering Large language models
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

86

Forks

7

Language

Python

License

Apache-2.0

Last pushed

Aug 19, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/AIFEG/BenchLMM"

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