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

SelfCheckGPT

Benchmarks

Task NameDataset NameSOTA ResultTrend
Hallucination CorrectionSelfCheckGPT
Reliability Score62,500,000,000
64
Hallucination DetectionSelfCheckGPT
AUCPR43
18
Hallucination DetectionSelfCheckGPT
AUROC67.3
18
Showing 3 of 3 rows