List your company
ResearchFormerly WizWand

Hallucination Detection on RAGTruth RT-D2T 1.0 (test)

0.7383F1 ScoreWavelet-high
0.4687320.5387160.60870.678684Feb 20, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Wavelet-highModel=LLaMA-13B2026.020.73830.8932
Fourier-highModel=LLaMA-13B2026.020.72780.8825
Fourier-highModel=Mistral-7B2026.020.72670.8863
LaplacianModel=Mistral-7B2026.020.72550.8849
LaplacianModel=LLaMA-13B2026.020.71940.8796
Lookback-lensModel=Mistral-7B2026.020.71510.8845
Lookback-lensModel=LLaMA-13B2026.020.71370.8766
Wavelet-highModel=Mistral-7B2026.020.71360.8829
Attn-entropyModel=Mistral-7B2026.020.70490.8734
Attn-entropyModel=LLaMA-13B2026.020.70340.8717
SelfcheckGPTModel=Mistral-7B2026.020.69580.8353
Wavelet-highModel=LLaMA-7B2026.020.68980.8569
LaplacianModel=LLaMA-7B2026.020.68780.8519
Fourier-highModel=LLaMA-7B2026.020.6870.8595
SelfcheckGPTModel=LLaMA-7B2026.020.65520.8026
SelfcheckGPTModel=LLaMA-13B2026.020.63460.7989
Attn-varianceModel=Mistral-7B2026.020.62180.7021
Lookback-lensModel=LLaMA-7B2026.020.61750.8442
EigenScoreModel=LLaMA-13B2026.020.60630.6582
RedeepModel=Mistral-7B2026.020.60130.6648
Attn-entropyModel=LLaMA-7B2026.020.60110.8368
Attn-varianceModel=LLaMA-13B2026.020.59920.7177
RefCheckerModel=Mistral-7B2026.020.59070.6287
RefCheckerModel=LLaMA-13B2026.020.57840.6381
RefCheckerModel=LLaMA-7B2026.020.57130.6349
EigenScoreModel=Mistral-7B2026.020.54490.6507
RedeepModel=LLaMA-13B2026.020.52730.6449
EigenScoreModel=LLaMA-7B2026.020.52560.5297
Attn-varianceModel=LLaMA-7B2026.020.48390.589
RedeepModel=LLaMA-7B2026.020.47910.396