Text-oriented Visual Question Answering on TextCaps
144.9CIDErCogVLM
Evaluation results
| Method | Links | |
|---|---|---|
| CogVLMResolution=4902024.04 | 144.9 | |
| HRVDAResolution=15362024.04 | 125.3 | |
| UReaderResolution=2242024.04 | 118.4 | |
| mPLUG-DocResolution=2242024.04 | 111.9 | |
| SeRumResolution=1280, Fine-tuning=true2024.04 | 101.4 | |
| DonutResolution=1280, Fine-tuning=true2024.04 | 74.4 | |
| Qwen-VLResolution=448, Open-source evaluation=true2024.04 | 20.25 |