Short Video Question Answering on TOMATO (test)
42AccuracyTarsier2-7B
Evaluation results
| Method | Links | |
|---|---|---|
| Tarsier2-7BParameters=7B2025.01 | 42 | |
| Qwen2-VL-72BModel Group=Open-source (>10B), Parameters=72B2025.01 | 37.9 | |
| GPT-4oModel Group=Proprietary2025.01 | 37.7 | |
| Gemini-1.5-ProModel Group=Proprietary2025.01 | 36.1 | |
| Tarsier-34BModel Group=Open-source (>10B), Parameters=34B2025.01 | 34.3 | |
| Qwen2-VL-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 31.5 | |
| Previous SOTACitation=[106]2025.01 | 31.5 | |
| LLaVA-OV-72BModel Group=Open-source (>10B), Parameters=72B2025.01 | 28.6 | |
| Tarsier-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 28.6 | |
| LLaVA-Video-72BModel Group=Open-source (>10B), Parameters=72B2025.01 | 28.2 | |
| LLaVA-OV-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 25.5 | |
| LLaVA-Video-7BModel Group=Open-source (<=10B), Parameters=7B2025.01 | 24.9 |