Video Question Answering On Next Qa Efficient
Metrics
1:1 Accuracy
Results
Performance results of various models on this benchmark
| Paper Title | ||
|---|---|---|
| ViLA (3B, 4 frames) | 74.4 | ViLA: Efficient Video-Language Alignment for Video Question Answering |
| SeViLA (4 frames) | 73.8 | Self-Chained Image-Language Model for Video Localization and Question Answering |
0 of 2 row(s) selected.