| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| UltraFeedback (UF) | ANCHOR | AOG7.93 | 16 | 8d ago | |
| Anthropic Helpful (AH) | AOG9.95 | 16 | 8d ago | ||
| HelpSteer3 (H3) v1 (test) | ANCHOR | AOG8.96 | 16 | 8d ago | |
| RAIL | ANCHOR | AOG4.62 | 8 | 8d ago | |
| PKU-SafeRLHF | ANCHOR | AOG2.46 | 8 | 8d ago | |
| Anthropic Harmless (AHar) | ANCHOR | AOG4.62 | 8 | 8d ago |