Video Quiz Evaluation on SciVidEval
99.5VLM-as-Judge ScoreSource Paper
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Source Paper2025.09 | 99.5 | 95 | |
| VideoAgentBackbone=Gemini-2.5 Pro, Summary Mode=Default Sequential Summary2025.09 | 99.5 | 87.5 | |
| NotebookLM2025.09 | 99 | 86 | |
| VideoAgentBackbone=GPT-4o, Summary Mode=Default Sequential Summary2025.09 | 99 | — | |
| AutoSlides2025.09 | 98.5 | 78 | |
| VideoAgentBackbone=Qwen-2.5VL-7B, Summary Mode=Default Sequential Summary2025.09 | 98 | — | |
| Author2025.09 | 97 | 90 | |
| PresentAgent2025.09 | 97 | 75 | |
| LunWenShuo2025.09 | 89.5 | 79.5 | |
| Pictory2025.09 | 85.5 | 74 |