Visual Instruction Evaluation on LRV-Instruction
6.58Accuracy (0-10)Finetuned mPLUG-Owl-7B
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Finetuned mPLUG-Owl-7BModel Scale=7B, Fine-tuning status=Finetuned2023.06 | 6.58 | 8.46 | 3.48 | 3.58 | 3.33 | |
| InstructBLIPModel Scale=7B2023.06 | 5.93 | 7.34 | 3 | 2.48 | 2.94 | |
| mPLUG-OwlModel Scale=7B2023.06 | 4.84 | 6.35 | 2.9 | 2.27 | 2.91 | |
| LLaVAModel Scale=7B2023.06 | 4.36 | 6.11 | 2.87 | 2.07 | 2.89 | |
| MiniGPT4Model Scale=7B2023.06 | 4.14 | 5.81 | 2.61 | 2.23 | 2.58 | |
| MMGPTModel Scale=7B2023.06 | 0.91 | 1.79 | 1.9 | 1.05 | 1.38 |