List your company

Hallucination Detection on AG News (AUROC)

0.852AUROCUQ_ICL
0.51920.60560.6920.7784Apr 28, 2026

Evaluation results

MethodLinks
UQ_ICLModel=LLaMA2-13B2026.040.852
Function VectorModel=LLaMA2-7B2026.040.85
Function VectorModel=LLaMA2-13B2026.040.848
Self-FVModel=LLaMA2-7B2026.040.845
Self-FVModel=LLaMA2-13B2026.040.845
Total EntropyModel=LLaMA2-7B2026.040.843
MaxProbModel=LLaMA2-7B2026.040.837
Total EntropyModel=LLaMA2-13B2026.040.832
UQ_ICLModel=LLaMA2-7B2026.040.805
Semantic EntropyModel=LLaMA2-13B2026.040.598
Lookback LensModel=LLaMA2-7B2026.040.572
Semantic EntropyModel=LLaMA2-7B2026.040.532