StereoSet
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
StereoSet
63.17Overall SS
25
StereoSet Gender
99.59LMS Score
24
Stereoset
67.2Accuracy
14
StereoSet (test)
93Macro F1 Score
12
StereoSet (dev)
84.172Overall LMS Score
12
StereoSet (SS)
41StereoSet Score (SS)
10
StereoSet Explicit Preference (test)
79.1Preference Score
8
StereoSet Implicit Preference (test)
0.474Pref Score
8
StereoSet (test)
77.12Gender SS
8
StereoSet Overall
91.05LMS
8
StereoSet n=237
93.4Accuracy
5
StereoSet (test)
0.854LMS Score
5
StereoSet intrasentence
67.34Gender SS
3
StereoSet Race
77LMS
2
StereoSet Religion
84LMS
2
StereoSet Profession
78.4LMS
2