Visual Question Answering On Grit
Metrics
VQA (ablation)
VQA (test)
Results
Performance results of various models on this benchmark
| Paper Title | |||
|---|---|---|---|
| Unified-IOXL | 74.5 | 74.5 | Unified-IO: A Unified Model for Vision, Language, and Multi-Modal Tasks |
| GPV-2 | 63.5 | 63.2 | Webly Supervised Concept Expansion for General Purpose Vision Models |
0 of 2 row(s) selected.