Visual Question Answering on InfoHaystack-1000
60AccuracyQwen2-VL + V-RAG
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2-VL + V-RAGFramework=V-RAG, Resolution=Low-resolution2024.11 | 60 | |
| Qwen2-VL-f.t. + V-RAGFramework=V-RAG, Resolution=Low-resolution, Adaptation=LoRA2024.11 | 60 | |
| GPT-4o + V-RAGFramework=V-RAG, Resolution=Low-resolution2024.11 | 56.77 | |
| Gemini + V-RAGFramework=V-RAG2024.11 | 47.1 | |
| LLaVA-OV + V-RAGFramework=V-RAG2024.11 | 36.77 | |
| MIRAGE2024.11 | 6.45 |