Visual Question Answering on Personalized Visual Question Answering 1.0 (test)
93.6AccuracyGPT-4V + Prompt
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4V + PromptPrompt Type=Human, Number of Tokens=~162024.06 | 93.6 | |
| GPT-4V + PromptPrompt Type=Text, Number of Tokens=~162024.06 | 93.2 | |
| Yo'LLaVAPrompt Type=Learnable, Number of Tokens=162024.06 | 92.9 | |
| LLaVA + PromptPrompt Type=Text, Number of Tokens=~162024.06 | 92.5 | |
| LLaVA + PromptPrompt Type=Human, Number of Tokens=~162024.06 | 91.3 | |
| LLaVAPrompt Type=0, Number of Tokens=02024.06 | 89.9 | |
| GPT-4V + PromptPrompt Type=Image, Number of Tokens=~5k2024.06 | 88.7 | |
| LLaVA + PromptPrompt Type=Text, Number of Tokens=~1.3k2024.06 | 88.3 | |
| GPT-4V + PromptPrompt Type=Image, Number of Tokens=~1k2024.06 | 86.6 |