Video Understanding on MLVU (MLVU, Avg)
73.83MLVU ScoreFLoC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FLoCModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 73.83 | 70.63 | |
| DivPruneModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 73.37 | 69.32 | |
| TS-LLaVAModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 73.09 | 69.44 | |
| FLoCModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 72.63 | 67.03 | |
| RandomModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 72.49 | 68.94 | |
| TS-LLaVAModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 72.4 | 66.52 | |
| DyCokeModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 72.26 | 69.04 | |
| LongVUModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 72.22 | 69.07 | |
| -Model=Qwen2.5-VL-32B, Max Frames=7682025.10 | 71.57 | 68.19 | |
| DyCokeModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 71.3 | 66.69 | |
| LongVUModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 71.02 | 66.27 | |
| RandomModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 70.52 | 65.57 | |
| -Model=Qwen2.5-VL-7B, Max Frames=7682025.10 | 70.31 | 65.82 | |
| DivPruneModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 70.19 | 65.81 | |
| PruneVidModel=Qwen2.5-VL-32B, Max Frames=72002025.10 | 70.19 | 67.23 | |
| PruneVidModel=Qwen2.5-VL-7B, Max Frames=72002025.10 | 68.63 | 64.68 |