V*Bench: A Visual QA Benchmark with Detailed High-resolution Images

Active

V*Bench is a visual question & answer benchmark that evaluates MLLMs in their ability to process high-resolution and visually crowded images to find and focus on small details.

Publisher: University of California, San Diego
Domain: Multimodal
License: mit
Published: Oct 2024
Notable for: Benchmark for evaluating Multimodal.
Canonical: github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/vstar_bench

Cite

Notes

Only stored in your browser.

Attribution

README: github.com/UKGovernmentBEIS/inspect_evals/blob/main/src/inspect_evals/vstar_bench/eval.yamlMIT

Attribution policy →

FAQ

What is V*Bench: A Visual QA Benchmark with Detailed High-resolution Images?: V*Bench is a visual question & answer benchmark that evaluates MLLMs in their ability to process high-resolution and visually crowded images to find and focus on small details.
What license is V*Bench: A Visual QA Benchmark with Detailed High-resolution Images under?: V*Bench: A Visual QA Benchmark with Detailed High-resolution Images is available under mit.