Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Hallucination Detection on RAGTruth RT-D2T 1.0 (test)

0.7383F1 Score

Wavelet-high

0.4687320.5387160.60870.678684Feb 20, 2026
Updated 5mo ago

Evaluation Results

MethodLinks
2026.02
0.73830.8932
2026.02
0.72780.8825
2026.02
0.72670.8863
2026.02
0.72550.8849
2026.02
0.71940.8796
2026.02
0.71510.8845
2026.02
0.71370.8766
2026.02
0.71360.8829
2026.02
0.70490.8734
2026.02
0.70340.8717
2026.02
0.69580.8353
2026.02
0.68980.8569
2026.02
0.68780.8519
2026.02
0.6870.8595
2026.02
0.65520.8026
2026.02
0.63460.7989
2026.02
0.62180.7021
2026.02
0.61750.8442
2026.02
0.60630.6582
2026.02
0.60130.6648
2026.02
0.60110.8368
2026.02
0.59920.7177
2026.02
0.59070.6287
2026.02
0.57840.6381
2026.02
0.57130.6349
2026.02
0.54490.6507
2026.02
0.52730.6449
2026.02
0.52560.5297
2026.02
0.48390.589
2026.02
0.47910.396