Sycophancy Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.123Total Sycophancy Score
32
Apr 21, 2026
1.6Sycophancy
24
Jun 11, 2026
4.7Sycophancy Rate
24
Jun 11, 2026
8.14PSS
14
Apr 8, 2026
0.112PSS
14
Apr 8, 2026
96A/B Accuracy
12
May 19, 2026
99.34Sycophancy Preference
10
Feb 26, 2026
92.18Sycophantic Preference (%)
10
Feb 26, 2026
49.25Sycophancy Preference
10
Feb 26, 2026
48.15Syc Score
10
Feb 26, 2026
1.21Pickside Score
10
Feb 26, 2026
13.3BRR
9
May 22, 2026
58.04Strong Bias
9
May 1, 2026
54.2Sycophancy Rate
9
Apr 2, 2026
106Correct Count
8
Jun 9, 2026
2.41Mean Sycophancy
8
Jun 5, 2026
0.481S (PD, L)
6
Apr 6, 2026
0.54Sycophancy Score (S) PD-L
6
Apr 6, 2026
88.2Observed Praise Rate
4
Jun 8, 2026
2.64Control Mean
3
Jun 5, 2026
1.83Mean Sycophancy Score
3
Jun 5, 2026
64.84Strong Bias
3
May 1, 2026
1Sycophantic Response Count
3
Apr 2, 2026
2Sycophantic Responses Count
3
Apr 2, 2026
4Sycophancy Prevalence Score
3
Feb 26, 2026