Model Fable Fairytale Science History Folklore Movie Average BLIP-2 (zero-shot) 12.5 10.9 7.5 7.5 16.1 - 10.7 InstructBLIP (zero-shot) 55.0 28.3 55.0 57.5 29.0 - 45.2 mPLUG-Owl (zero-shot) 47.5 41.3 62.5 45.0 41.9 - 47.7 mPLUG-Owl2 (zero-shot) 47.5 65.2 80.0 67.5 45.2 - 61.9 LLaVA-v1.5 (zero-shot) 60.0 58.7 70.0 75.0 41.9 - 61.9 LLaVA-v1.6 (zero-shot) 55.0 52.2 70.0 45.0 48.4 - 54.3 MMICL (zero-shot) 27.5 23.9 35.0 32.5 25.8 - 28.9 OpenFlamingo (zero-shot) 10.0 0.0 12.5 5.0 0.0 - 5.6 Otter (zero-shot) 15.2 18.2 20.5 11.2 17.4 - 16.5 GPT-4V (zero-shot) 80.0 84.8 95.0 82.5 83.9 - 85.3 MMICL (few-shot) 22.5 23.9 22.5 20.0 38.7 - 24.9 OpenFlamingo (few-shot) 12.5 26.1 17.5 20.0 12.9 - 18.3 Otter (few-shot) 13.3 17.5 17.2 16.5 22.3 - 17.3 GPT-4V (few-shot) 85.0 84.8 90.0 87.5 87.1 - 86.8 MMICL (CoCoT) 20.0 19.6 35.0 20.0 12.9 - 21.8 OpenFlamingo (CoCoT) 0.0 0.0 0.0 0.0 3.2 - 0.5 Otter (CoCoT) 8.2 12.5 5.0 25.5 5.0 - 11.2 GPT-4V (CoCoT) 87.5 89.1 95.0 87.5 83.9 - 88.8 Human 98.1 97.7 95.5 96.3 95.5 - 96.6