Visual Question Answering on Multi-Concept Personalized VLM Benchmark
69.8VQA AccuracyGPT-4V + Prompt
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4V + PromptPrompting=true2024.11 | 69.8 | |
| MC-LLaVA2024.11 | 67.9 | |
| Qwen2.5-VL-7B + PromptPrompting=true2024.11 | 51.8 |
| Method | Links | |
|---|---|---|
| GPT-4V + PromptPrompting=true2024.11 | 69.8 | |
| MC-LLaVA2024.11 | 67.9 | |
| Qwen2.5-VL-7B + PromptPrompting=true2024.11 | 51.8 |