Leaderboards
We host results only for Closed-Book methods that have been finetuned on only In-Domain data.Submissions
To add your system to our leaderboards, please see our submission instructions.
Closed-Book: In-Domain
| Rank | System | Dev Set | Test Set | Contrastive Set |
|---|---|---|---|---|
| - | Human (ensembled) | 99.0 | - | 99.0 |
| - | Human (averaged) | 96.3 | - | 92.2 |
| 1 | RACo-Large | 88.2 | 88.6 | 74.4 |
| 2 | T5-3B | 85.6 | 85.1 | 70.0 |
| 3 | RoBERTa-Large | 80.6 | 80.3 | 61.5 |
| 4 | RoBERTa-Base | 72.2 | 71.6 | 56.0 |