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

Hallucination Detection on GSM8K

90.37AUROC

ARS (CCS)

41.38654.10366.8279.537Feb 6, 2025Apr 18, 2025Jun 29, 2025Sep 9, 2025Nov 19, 2025Jan 30, 2026Apr 12, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
90.37
2026.01
89.88
2026.04
85.4
2026.04
84.5
2026.04
84.5
2026.04
84.3
2026.04
84.1
2026.01
83.78
2026.04
83.5
2026.01
83.15
2025.02
82.7
2026.04
82.6
2026.04
82.5
2026.04
82.4
2026.04
81.9
2026.04
81.6
2026.04
81.6
2026.04
81.3
2026.04
81.2
2026.04
81.1
2026.02
80.6
2026.04
80.4
2026.04
80.3
2026.04
80.2
2026.04
79.7
2026.04
79.7
2026.04
79.4
2025.02
79.25
2025.02
78.5
2026.01
78.29
2026.04
78.2
2026.04
78
2026.01
77.62
2026.04
77.5
2025.02
77.2
2026.02
77.19
2026.02
77.19
2026.02
76.86
2025.02
76.53
2026.02
76.36
2026.02
76.32
2026.02
76.22
2025.02
76.14
2025.02
75.85
2026.02
75.83
2026.02
75.75
2026.02
75.59
2026.02
75.5
2026.04
75.2
2026.02
75.13
2026.02
74.99
2026.01
74.72
2026.02
74.38
2026.04
74.3
2026.02
74.29
2026.04
74.1
2026.02
73.9
2026.02
73.66
2026.02
73.29
2026.02
73.01
2026.01
72.55
2026.01
72.51
2025.02
72.51
2026.02
72.02
2025.02
71.56
2026.02
71.54
2026.04
71.2
2026.02
70.88
2026.02
70.68
2026.01
70.38
2026.02
70.31
2026.02
69.63
2026.01
68.59
2026.02
66.83
2026.01
66.38
2025.02
65.86
2026.02
64.4
2026.01
61.98
2026.01
61.67
2026.01
60.8
2026.02
60.69
2026.01
58.48
2026.01
57.6
2025.02
57.11
2026.01
56.01
2026.01
54.33
2025.02
51.36
2026.02
51.02
2026.01
50.58
2026.02
50.12
2026.02
50
2026.01
50
2026.01
43.27