| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Anthropic Model-Written Evaluations (MWE) In-distribution | L2S w/ W2S | Steerability2.363 | 8 | 12d ago | |
| BASE→USER.NEG (out-of-distribution) | L2S | Steerability2.195 | 8 | 12d ago | |
| BASE→USER.POS (out-of-distribution) | L2S | Steerability2.535 | 8 | 12d ago | |
| BASE→SYS.NEG (out-of-distribution) | L2S | Steerability2.944 | 8 | 12d ago | |
| BASE→SYS.POS (out-of-distribution) | L2S | Steerability2.802 | 8 | 12d ago |
