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

PsiloQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Hallucination DetectionPsiloQA
AUROC92.64
56
Hallucination DetectionPsiloQA
Accuracy88.43
14
Span-level Hallucination DetectionPsiloQA (test)
F1 Score73.2
3
Hallucination DetectionPsiloQA 14-lang IoU
IoU68.9
3
Hallucination DetectionPsiloQA EN (IoU)
IoU72.4
3
Showing 5 of 5 rows