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

Hallucination prediction on HotpotQA (original)

83.39AUROC

Conf + Probe (SCAO)

72.927675.643878.3681.0762Sep 18, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
83.3915.02
2025.09
81.0812.71
2025.09
80.5812.21
2025.09
77.7-
2025.09
77.3313.55
2025.09
77.2813.5
2025.09
74.88-
2025.09
74.4110.63
2025.09
74.13-
2025.09
73.33-