Video Question Answering on VideoHallucer
88.3Basic ScoreLLAVA-OV-72B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLAVA-OV-72BModel Type=Open-Source, Model Scale=>10B2025.01 | 88.3 | 62.6 | 55.2 | |
| LLaVA-Video-72BModel Type=Open-Source, Model Scale=>10B2025.01 | 88.2 | 73.5 | 64.6 | |
| Qwen2-VL-72BModel Type=Open-Source, Model Scale=>10B2025.01 | 87.1 | 79.4 | 70.2 | |
| Tarsier2-7BModel Type=Open-Source, Model Scale=≤10B2025.01 | 86.5 | 78.3 | 67 | |
| Qwen2-VL-7BModel Type=Open-Source, Model Scale=≤10B2025.01 | 85 | 70.8 | 59.3 | |
| Tarsier-34BModel Type=Open-Source, Model Scale=>10B2025.01 | 84.8 | 80 | 67.7 | |
| Gemini-1.5-ProModel Type=Proprietary2025.01 | 83.6 | 42.3 | 37.8 | |
| InternVL2.5-78BModel Type=Open-Source, Model Scale=>10B2025.01 | 82.5 | 82.5 | 67.8 | |
| LLaVA-Video-7BModel Type=Open-Source, Model Scale=≤10B2025.01 | 82.4 | 70.6 | 56 | |
| LLAVA-OV-7BModel Type=Open-Source, Model Scale=≤10B2025.01 | 81.1 | 69.6 | 53.8 | |
| Tarsier-7BModel Type=Open-Source, Model Scale=≤10B2025.01 | 76.4 | 60.8 | 41.4 | |
| GPT-4oModel Type=Proprietary2025.01 | 75.1 | 74.2 | 53.3 | |
| InternVL2.5-8BModel Type=Open-Source, Model Scale=≤10B2025.01 | 72.7 | 78.3 | 53.6 |