outputs_vlmeval/llava-qwen2-instruct-llm-ve/llava-qwen2-instruct-llm-ve_MMStar_acc.csv

"split","Overall","coarse perception","fine-grained perception","instance reasoning","logical reasoning","math","science & technology"
"none","0.36133333333333334","0.524","0.292","0.496","0.316","0.312","0.228"