Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Truthfulness Evaluation on TruthfulQA (test)

54.95MC1

PromptCD

27.07834.31441.5548.786May 15, 2024Aug 31, 2024Dec 17, 2024Apr 5, 2025Jul 22, 2025Nov 7, 2025Feb 24, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
54.9571.13---45.93
2026.02
50.7969.58---44.24
2026.02
49.7968.89---39.4
2026.02
49.2169.59---43.83
2026.02
48.5966.24---37.45
2026.02
47.2467.61---36.9
2026.02
46.8865.15---35.22
2026.02
45.4168.43---40.35
2026.02
44.9270.35---39.82
2026.02
42.5964.05---33.89
2026.02
41.9861.32---33.11
2026.02
41.3769.17---37.88
2026.02
41.1267.99---37.51
2024.05
39.4157.1568.0550.6733.66-
2026.02
39.1668.27---35.69
2024.05
38.3155.2770.3848.9635.25-
2024.05
37.8757.77----
2026.02
37.0863.83---32.92
2024.05
37.0154.66----
2024.05
36.9654.6869.447.3633.29-
2024.05
35.7454.6191.0648.5942.59-
2026.02
35.7458.76---30.68
2024.05
35.0154.2492.4147.4942.35-
2026.02
33.6651.29---24.91
2024.05
32.9760.84----
2026.02
32.6861.12---29.86
2026.02
32.1950.38---25.06
2026.02
31.9548.83---23.92
2024.05
28.3943.42----
2024.05
28.1554.87----
2024.05
--69.447.3633.29-
2024.05
--76.557.0344.55-
2024.05
--80.5462.350.31-
2024.05
--68.0550.6733.66-
2024.05
--77.7257.2844.56-
2024.05
--82.0163.0451.3-