Legal Reasoning on LegalBench (Issue-spotting)
Metric: Balanced Accuracy (higher is better)
LeaderboardDataset
Results
Submit a result| # | Model↕ | Balanced Accuracy▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | GPT-4 | 82.9 | No | - | - | - |
| 2 | GPT-3.5 | 60.9 | No | - | - | - |
| 3 | Claude-1 | 58.1 | No | - | - | - |