List your company

Hallucination Detection on WNMCQ1 (AUROC)

91.2AUROCUQ_ICL
55.8465.0274.283.38Apr 28, 2026

Evaluation results

MethodLinks
UQ_ICLModel=LLaMA2-13B2026.0491.2
Semantic EntropyModel=LLaMA2-13B2026.0490.4
Self-FVModel=LLaMA2-7B2026.0489.93
Function VectorModel=LLaMA2-13B2026.0489.9
Function VectorModel=LLaMA2-7B2026.0489.89
Self-FVModel=LLaMA2-13B2026.0488.9
UQ_ICLModel=LLaMA2-7B2026.0488.4
MaxProbModel=LLaMA2-7B2026.0487.5
Total EntropyModel=LLaMA2-13B2026.0487.3
Total EntropyModel=LLaMA2-7B2026.0487.2
Semantic EntropyModel=LLaMA2-7B2026.0485.9
Lookback LensModel=LLaMA2-7B2026.0457.2