C-Eval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
C-Eval
92.5Accuracy
68
C-Eval
92.5Accuracy
46
C-Eval (test)
93.02Natural Sciences Score
36
C-Eval
89Accuracy
30
C-Eval
87.7C-Eval Score
29
C-Eval
81.4Accuracy
28
C-Eval
91.82Accuracy
19
C-Eval
77.1Score
18
C-Eval
0.682C-Eval Knowledge Accuracy
18
C-Eval (val)
78.68Accuracy
18
C-EVAL
88.12Score
17
C-Eval (val)
34.32Accuracy
15
C-Eval
78.5Overall Score
14
C-Eval (test)
71.5Accuracy
14
C-Eval
42.79Accuracy
13
C-Eval (test)
71.8Accuracy
13
C-Eval (val)
83C-Eval 0-shot Score
12
C-Eval 1.0 (val)
78.68Accuracy
12
C-Eval
90.2Score
7
C-Eval (test)
86Accuracy
7
C-Eval
82.7Accuracy
6
C-Eval
91.8Exact Match
6
C-Eval
83.28Accuracy
4
C-Eval
92.5Exact Match
4
C-Eval
78Accuracy
3