IFEval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
IFEval Galician (GL)
79Strict Accuracy
13
IFEval Basque (EU)
62.81Strict Accuracy
13
IFEval
85.1ROCG
12
IFEval
94.3Accuracy (%)
12
IFEval EN
87.64Score
12
IFEval
89.8Score
12
IFEval
84.53Accuracy
11
IFEval English (test)
93.16IFEval Accuracy
10
IFEval et
61.4Instruction Level Strict Accuracy
10
IFEval standard (test)
100Punctuation Score
10
IFEval
86.3IFEval Score
10
IFEval
45.2IFEval Delta
9
IFEval
0False Alarm Rate
9
IFEval
19.6Accuracy (Prompt)
8
IFEval
82.81Loose Accuracy (Prompt-level)
8
IFEval latest (test)
59.52IF-pS
8
IFEVAL MINCE Calibration n=250
89.53Accuracy
7
IFEVAL Full Standard
88.97Accuracy
7
IFEval (test)
75.23Prompt Loose Accuracy
7
IFEval English
90.9Accuracy
6
IFEval Korean
78.3Accuracy
6
IFEval
89.4IFEval avg@4
6
IFEVAL
36.13IFEVAL Score
6
IFEval Synthetic
6.58Personalization Score
6
IFEval-TH
99.4IFEval-TH Score
6