Human Evaluation Dataset
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Humorous Image Captioning | Human Evaluation Dataset (test) | Elo Rating1,554.6 | 9 | |
| Human Evaluation of Code-Switching Generation | Human Evaluation Dataset 150 samples from 5 sources | Syntactic Score4.47 | 5 | |
| Claim Verification | Human evaluation dataset 400 claims | Accuracy0.937 | 3 |
Showing 3 of 3 rows