Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Response-level Hallucination Detection on RAGTruth QA
Loading...
91.89
AUROC
LettuceDetect-L
51.7044
62.1372
72.57
83.0028
Apr 17, 2026
AUROC
Updated 1mo ago
Evaluation Results
Method
Method
Links
AUROC
LettuceDetect-L
Detector category=Blac...
2026.04
91.89
HDM-2-3B
Detector category=Blac...
2026.04
87.95
LookbackLens
Detector category=Whit...
2026.04
87.54
RAGognizer
Detector category=Whit...
2026.04
87.43
HallucinationProbes
Detector category=Whit...
2026.04
78.08
HHEM-2.1-Open
Detector category=Blac...
2026.04
71.7
Semantic Entropy
Detector category=Whit...
2026.04
67.98
MiniCheck-7B
Detector category=Blac...
2026.04
63.1
SelfCheckGPT (NLI)
Detector category=Blac...
2026.04
61.13
INSIDE (EigenScore)
Detector category=Whit...
2026.04
60.96
DeBERTa-v3 (Con.)
Detector category=Blac...
2026.04
56.69
HaloScope
Detector category=Whit...
2026.04
53.54
DeBERTa-v3 (Ent.)
Detector category=Blac...
2026.04
53.25
Feedback
Search any
task
Search any
task