Video Question Answering on MINERVA
65.6Accuracy (%)AVP
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| AVP2025.12 | 65.6 | 0.62 | 0.82 | 0.97 | 0.93 | 0.84 | |
| Gemini 2.5 Pro2025.12 | 61.8 | 0.6 | 0.62 | 0.97 | 0.78 | 0.74 | |
| Gemini 2.0 Flash2025.12 | 53.5 | 0.62 | 0.75 | 0.83 | 0.82 | 0.75 | |
| GPT-4o2025.12 | 45.5 | 0.57 | 0.67 | 0.77 | 0.79 | 0.7 | |
| OpenAI o12025.12 | 43.5 | 0.52 | 0.52 | 0.86 | 0.88 | 0.69 |