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

Hallucination Detection on MMLU

87.48AUPRC

Lyapunov Probes

21.23238.43155.6372.829Mar 6, 2026Mar 13, 2026Mar 20, 2026Mar 27, 2026Apr 3, 2026Apr 10, 2026Apr 17, 2026
Updated 28d ago

Evaluation Results

MethodLinks
2026.03
87.48----
2026.03
87.35----
2026.04
81.6554.0796.3--
2026.03
80----
2026.03
79.26----
2026.04
78.4858.9492.77--
2026.04
78.4858.9492.77--
2026.04
75.7757.3196.51--
2026.04
75.7757.3196.51--
2026.04
75.5453.6496.1--
2026.04
74.8846.69100--
2026.04
74.8846.69100--
2026.04
74.2847.7596.15--
2026.04
74.2847.7596.15--
2026.04
73.7849.594.16--
2026.04
73.7746.6497.4--
2026.04
73.5549.2790.32--
2026.04
73.4349.6292.86--
2026.04
73.449.8494.16--
2026.04
73.3849.3994.81--
2026.04
73.3649.4893.51--
2026.04
72.8746.4298.05--
2026.04
72.6751.2596.5--
2026.04
72.6751.2596.5--
2026.04
72.6454.4189.47--
2026.04
72.6454.4189.47--
2026.04
72.6346.3395.38--
2026.04
72.6346.3395.38--
2026.04
72.5550.3794.29--
2026.04
72.5451.2195.1--
2026.04
72.5451.2195.1--
2026.04
72.4150.6994.87--
2026.04
72.4150.69---
2026.04
72.3350.5296.27--
2026.04
72.3350.5296.27--
2026.04
72.2550.495.34--
2026.04
72.2550.495.34--
2026.04
72.2350.5795.1--
2026.04
72.2350.5795.1--
2026.04
72.0550.3795.34--
2026.04
72.0550.3795.34--
2026.03
71.7----
2026.04
70.6549.6296.04--
2026.04
70.6549.6296.04--
2026.04
69.7838.4991.43--
2026.03
67.61----
2026.04
66.548.7597.06--
2026.03
65.43----
2026.04
64.2147.0191.18--
2026.03
57.48----
2026.03
54.09----
2026.03
52.43----
2026.03
33.79----
2026.03
33.6----
2026.03
30.77----
2026.03
29.41----
2026.03
28.99----
2026.03
25.11----
2026.03
24.98----
2026.03
24.79----
2026.03
23.9----
2026.03
23.78----
2026.05
---57.2965.13
2026.05
---59.7863.56
2026.05
---57.8863.47
2026.05
---69.9670.75
2026.05
---71.7773.25
2026.05
---65.2766.07
2026.05
---66.7270.65
2026.05
---72.7473.96
2026.05
---70.9672.5
2026.05
---64.382.18
2026.05
---50.7569.7
2026.05
---59.3977.48
2026.05
---71.1979.62
2026.05
---71.481.75
2026.05
---62.4270.63
2026.05
---60.1872.33
2026.05
---68.9575.07
2026.05
---69.7379.62
2026.05
---65.1477.54
2026.05
---52.1776.86
2026.05
---56.9875.35
2026.05
---69.3878.38
2026.05
---69.678.99
2026.05
---53.1963.17
2026.05
---54.6274.36
2026.05
---68.1475.07
2026.05
---70.2979.79
2026.05
---42.2861.8
2026.05
---56.7360.4
2026.05
---55.8459.33
2026.05
---70.8971.36
2026.05
---70.9971.26
2026.05
---59.5660.26
2026.05
---60.9862.91
2026.05
---69.6169.68
2026.05
---71.2472.1