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

WorldValuesBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
User response predictionWorldValuesBench (held-out users)
Brier Score0.1347
56
Response PredictionWorldValuesBench (held-out)
Log Loss0.977
56
Ordinal PredictionWorldValuesBench (held-out users)
Ordinal MSE0.54
56
Cultural AlignmentWorldValuesBench (WVB)
Accuracy50.64
22
Computerized Adaptive TestingWorldValuesBench (test)
Inference Time (min)0.46
9
Showing 5 of 5 rows