Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SocialIQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Commonsense ReasoningSocialIQA
Accuracy88.1
116
Social Commonsense ReasoningSocialIQA
Accuracy87.11
100
Question AnsweringSocialIQA
Accuracy83.9
30
Commonsense Question AnsweringSocialIQA (SIQA) (val)
Accuracy70.7
24
Social Interaction Question AnsweringSocialIQA (test)
Accuracy75.49
18
Commonsense ReasoningSOCIALIQA (dev)
Accuracy73.8
11
Ranking correlation with full dataset evaluationSocialIQA
Kendall Correlation0.81
10
Scaling Law PredictionSocialIQA
MAE0.0088
7
Preference alignmentSocialIQA
Preference Alignment87.3
5
AdaptivitySocialIQA
Adaptivity75
4
Showing 10 of 10 rows