PolicyBench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
PolicyBench Overall Average
66.34Accuracy
11
PolicyBench Level 3 US
77Accuracy
11
PolicyBench Level 3 CN
80.34Accuracy
11
PolicyBench Level 2 (US)
68.95Accuracy
11
PolicyBench Level 2 (CN)
62.92Accuracy
11
PolicyBench Level 1 (US)
59.33Accuracy
11
PolicyBench Level 1 (CN)
62.02Accuracy
11
PolicyBench
64.34Accuracy
11
PolicyBench US
66.43Accuracy
11
PolicyBench Chinese
65.33Accuracy
11