Q-Future / Q-Bench

①[ICLR2024 Spotlight] (GPT-4V/Gemini-Pro/Qwen-VL-Plus+16 OS MLLMs) A benchmark for multi-modality LLMs (MLLMs) on low-level vision and visual quality assessment.
https://q-future.github.io/Q-Bench/
Other
251 stars 12 forks source link

Difference between llvisionqa_dev and llvisionqa_test #9

Closed Qiulin-W closed 11 months ago

Qiulin-W commented 11 months ago

Hi,

great work!

I'm wondering what is the difference between the two subset llvisionqa_dev and llvisionqa_dev? I can't find any clues from the paper and your project page.

Thanks!

teowu commented 11 months ago

We will provide an updated pre-print version soon discussing this split information.