Video Question Answering on Text-rich VQA
80.1ScoreFull Model
Evaluation results
| Method | Links | |
|---|---|---|
| Full ModelBackbone=LLaVA-OneVision, Sparsity Ratio=0%2026.04 | 80.1 | |
| LLM-StreamlineBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 73.6 | |
| TopoVLMBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 71.6 | |
| LLM-PrunerBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 69.8 | |
| TAMPBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 69.8 | |
| WandaBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 65.1 | |
| ECoFLaPBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 64.7 | |
| SparseGPTBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 63.6 | |
| OWLBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 60.1 | |
| MagnitudeBackbone=LLaVA-OneVision, Sparsity Ratio=60%2026.04 | 0 |