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

OOD Datasets

Benchmarks

Task NameDataset NameSOTA ResultTrend
OOD DetectionOOD Datasets Mean
FPR@9538.7
20
LLM RoutingOOD datasets (test)
Accuracy89
11
Few-shot classificationOOD Datasets (test)
Accuracy (Traffic Signs)91.33
11
Uncertainty Estimation23 OOD datasets
Mean AURC10.2
2
Showing 4 of 4 rows