List your company
ResearchFormerly WizWand

Hallucination Detection on RAGTruth RT-Summ 1.0 (test)

0.6966F1 ScoreSelfcheckGPT
0.4486640.5130320.57740.641768Feb 20, 2026
Updated 8mo ago

Evaluation results

MethodLinks
SelfcheckGPTModel=Mistral-7B2026.020.69660.7166
Wavelet-highModel=Mistral-7B2026.020.68490.8075
Fourier-highModel=Mistral-7B2026.020.67610.8037
Lookback-lensModel=Mistral-7B2026.020.67590.7954
LaplacianModel=Mistral-7B2026.020.67230.7978
RefCheckerModel=Mistral-7B2026.020.66640.6596
Attn-entropyModel=Mistral-7B2026.020.65510.7867
SelfcheckGPTModel=LLaMA-7B2026.020.63490.6674
RefCheckerModel=LLaMA-7B2026.020.60590.608
Wavelet-highModel=LLaMA-7B2026.020.59290.7165
Fourier-highModel=LLaMA-13B2026.020.59290.7362
RefCheckerModel=LLaMA-13B2026.020.59240.6346
RedeepModel=LLaMA-7B2026.020.58970.576
Fourier-highModel=LLaMA-7B2026.020.58750.7426
LaplacianModel=LLaMA-7B2026.020.57790.704
Lookback-lensModel=LLaMA-13B2026.020.56790.6702
Wavelet-highModel=LLaMA-13B2026.020.56510.7042
SelfcheckGPTModel=LLaMA-13B2026.020.55540.5909
LaplacianModel=LLaMA-13B2026.020.55480.67
Attn-entropyModel=LLaMA-13B2026.020.55270.6865
RedeepModel=Mistral-7B2026.020.53790.568
Lookback-lensModel=LLaMA-7B2026.020.53280.7156
EigenScoreModel=LLaMA-13B2026.020.51210.5097
EigenScoreModel=LLaMA-7B2026.020.50650.4989
Attn-entropyModel=LLaMA-7B2026.020.50310.6722
RedeepModel=LLaMA-13B2026.020.5010.5253
Attn-varianceModel=LLaMA-13B2026.020.49470.6629
Attn-varianceModel=LLaMA-7B2026.020.48860.6492
Attn-varianceModel=Mistral-7B2026.020.4840.5836
EigenScoreModel=Mistral-7B2026.020.45820.4973