| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Visio-linguistics Compositionality | Hard Positive Compositional Benchmark SWAP (test) | Test Accuracy (Original)66.1 | 3 | |
| Visio-linguistics Compositionality | Hard Positive Compositional Benchmark (test) | Original Test Accuracy68.3 | 3 |