Sycophancy Evaluation
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Sycophancy Evaluation Dataset
0.123Total Sycophancy Score
32
Sycophancy Evaluation Opinion
8.14PSS
14
Sycophancy Evaluation Factual
0.112PSS
14
Sycophancy Evaluation
13.3BRR
9
Sycophancy Evaluation Control condition
2.64Control Mean
3