| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| UltraFeedback 13B+ Models Holdout (test) | BENCHALIGN | Pairwise Accuracy (RM1_Honest)74.8 | 4 | 4d ago | |
| UltraFeedback 30B+ Models Holdout (test) | BENCHALIGN | Pairwise Acc (RM1_Honest)77.3 | 4 | 4d ago | |
| UltraFeedback 70B+ Models Holdout (test) | BENCHALIGN | Pairwise Acc (RM1_Honest)77.4 | 4 | 4d ago | |
| Helpsteer 13B+ Models Holdout (test) | BENCHALIGN | Acc_pair (RM1 Helpful)74.1 | 4 | 4d ago | |
| Helpsteer 30B+ Models Holdout (test) | BENCHALIGN | Pairwise Accuracy (RM1)76.5 | 4 | 4d ago | |
| Helpsteer 70B+ Models Holdout (test) | BENCHALIGN | Pairwise Acc (RM1)77.8 | 4 | 4d ago |