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

COMM

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationComm
Accuracy82.21
20
Commonsense ReasoningCOMM
Accuracy43.73
18
Fair ClassificationComm
Delta DP0.1731
15
Question-based GenerationCoMM validated (test)
Style Score7.35
5
Interleaved Image-Text GenerationCoMM (test)
Style Consistency9.22
4
Topological Preservation AnalysisComm. small
Kappa (FR)12.515
2
Private Information RetrievalComm 288GB
QPS544.6
2
Showing 7 of 7 rows