| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| FaithfulQA | REP | Sycophancy Score87.58 | 21 | 5mo ago | |
| Sycophancy dataset | System Prompt | Behavioral Alignment Score (Baseline)75.5 | 6 | 19d ago | |
| LFQA | Sycophancy (PD, L)0.276 | 6 | 3mo ago | ||
| MMLU | ACT | BRR Ratio3 | 5 | 1mo ago | |
| Sycophancy (Held-out Bias) | BRR Ratio100 | 5 | 1mo ago |