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

Com.

Benchmarks

Task NameDataset NameSOTA ResultTrend
Human Correlation for Image Captioning EvaluationCOM
Kendall's tau-c61.8
16
Commonsense ReasoningCom.
Vanilla Acc78.05
14
Time Series ClusteringCOM
Rand Index0.509
4
Showing 3 of 3 rows