Harmless-helpful
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Reward Model Controllability | Harmless-helpful | Kendall's Tau1 | 4 | |
| Generalization to Unseen Preferences | Harmless-helpful | Group 1 Score15.038 | 2 | |
| Controllability | Harmless-helpful Group 4 (unseen) | Kendall's Tau1 | 2 | |
| Controllability | Harmless-helpful Group 3 (unseen) | Kendall's Tau1 | 2 | |
| Controllability | Harmless-helpful Group 2 (unseen) | Kendall's tau1 | 2 | |
| Controllability | Harmless-helpful Group 1 (unseen) | Kendall's Tau1 | 2 |
Showing 6 of 6 rows