Image to Text on Oogiri English
66.92T1 ScoreCogVLM-17B + CLoT
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| CogVLM-17B + CLoTSize=7B2023.12 | 66.9 | 47.6 | 43.4 | 30.7 | 69.4 | 51.6 | |
| CogVLM-17B + AITSize=7B2023.12 | 57.4 | 37.4 | 33.5 | 21.8 | 64.6 | 42.9 | |
| CogVLM-17BSize=7B2023.12 | 37.6 | 26.4 | 18.3 | 2.5 | 64.6 | 29.9 | |
| mPLUG-OwlLLAMA2Size=7B2023.12 | 22.3 | 12.7 | 15 | 4.2 | 59.9 | 22.8 | |
| InstructionBLIPSize=13B2023.12 | 19.8 | 13.7 | 15.5 | 1.1 | 65.5 | 23.1 | |
| OtterSize=7B2023.12 | 15.8 | 9.9 | 8.5 | 7.1 | 61.3 | 20.5 |