Video Question Answering On Intentqa
Metriken
Accuarcy
CH
CW
TPu0026TN
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | |||||
|---|---|---|---|---|---|
| VideoChat2_HD_mistral | 83.4 | 90.0 | 84.0 | 77.3 | MVBench: A Comprehensive Multi-modal Video Understanding Benchmark |
| VideoChat2_mistral | 81.9 | 86.9 | 82.6 | 77.0 | MVBench: A Comprehensive Multi-modal Video Understanding Benchmark |
| Human | 78.5 | 80.2 | 77.8 | 79.1 | IntentQA: Context-aware Video Intent Reasoning |
| IntentQA | 57.6 | 65.5 | 58.4 | 50.5 | IntentQA: Context-aware Video Intent Reasoning |
| VGT | 51.3 | 56.0 | 51.4 | 47.6 | Video Graph Transformer for Video Question Answering |
| HQGA | 47.7 | 54.3 | 48.2 | 41.7 | Video as Conditional Graph Hierarchy for Multi-Granular Question Answering |
0 of 6 row(s) selected.