Benchmarks of LLAVA-CRITIC

[OctopusXXX]

LLAVA-CRITIC is an interesting job, thanks for sharing!

However, I had some questions about it. I noticed that the training data for the model had been shared, but the benchmark for In-domain Pointwise Scoring was not found, as mentioned in section 5.1 of the paper "we select 7 popular multimodal benchmarks and collect candidate responses from 13 commonly used LMMs alongside their GPT-4o evaluations, resulting in a total of 14174 examples"

Looking forward to your reply, thank you!