Visual Question Answering on InfoHaystack-100
67.1AccuracyQwen2-VL-f.t. + V-RAG
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2-VL-f.t. + V-RAGFramework=V-RAG, Resolution=Low-resolution, Adaptation=LoRA2024.11 | 67.1 | |
| Qwen2-VL + V-RAGFramework=V-RAG, Resolution=Low-resolution2024.11 | 65.81 | |
| GPT-4o + V-RAGFramework=V-RAG, Resolution=Low-resolution2024.11 | 65.16 | |
| Gemini + V-RAGFramework=V-RAG2024.11 | 57.42 | |
| LLaVA-OV + V-RAGFramework=V-RAG2024.11 | 43.22 | |
| Gemini2024.11 | 29.03 | |
| GPT-4oResolution=Low-resolution2024.11 | 23.87 | |
| Qwen2-VLResolution=Low-resolution2024.11 | 20 | |
| MIRAGE2024.11 | 7.74 |