List your company

Open-ended Generation on COCO 2014 (val)

8.58AccuracyLogicCheckGPT
3.23444.62226.017.3978Feb 18, 2024

Evaluation results

MethodLinks
LogicCheckGPTBase Model=QWEN-VL-Chat2024.028.589.96
vanillaBase Model=QWEN-VL-Chat2024.028.369.96
LogicCheckGPTBase Model=LLaVA-1.52024.026.57.64
LogicCheckGPTBase Model=MiniGPT-42024.026.028.38
vanillaBase Model=LLaVA-1.52024.025.227.24
vanillaBase Model=MiniGPT-42024.0257.96
LogicCheckGPTBase Model=mPLUG-Owl2024.024.328.74
vanillaBase Model=mPLUG-Owl2024.023.448.78