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

Object Hallucination Evaluation on POPE GQA

91.03Accuracy

Regular

67.182873.373979.56585.7561May 18, 2025Jul 24, 2025Sep 30, 2025Dec 7, 2025Feb 12, 2026Apr 21, 2026Jun 28, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2025.05
91.0390.75
2025.05
91.0390.67
2025.05
9190.7
2025.05
90.991.09
2025.05
90.189.52
2025.05
90.0389.34
2025.05
89.4388.9
2025.05
88.6387.86
2025.05
88.688.43
2025.05
88.4388.38
2025.05
88.488.33
2026.06
88.189
2025.05
87.988.94
2025.05
86.9785.34
2026.06
86.986.1
2025.05
86.8788.14
2025.05
86.886.25
2025.05
86.7787.55
2025.05
86.786.23
2025.05
86.5786.23
2025.05
86.5786.3
2025.05
86.5386.28
2026.01
86.4785.96
2025.05
86.185.41
2025.05
86.0387.52
2025.05
8687.33
2025.05
85.9785.43
2025.05
85.4787
2025.05
85.0783.55
2026.01
84.8484.48
2025.05
84.7783.06
2026.01
84.1383.78
2026.06
84.184.8
2025.05
83.7384.4
2025.05
83.7384.37
2026.01
83.5483.13
2026.01
83.4983.18
2025.05
83.1783.4
2025.05
82.9783.27
2025.05
82.9781.89
2026.01
82.6881.97
2025.05
82.5382.49
2025.05
82.4382.81
2026.01
82.3481.93
2026.01
81.8981.64
2025.05
81.8780.71
2025.05
81.7380.25
2026.01
81.4880.93
2025.05
81.4783.39
2025.05
81.3780.29
2026.01
81.3380.97
2026.01
81.1480.88
2025.05
81.1377.29
2026.01
80.4379.92
2025.05
79.9380.59
2025.05
79.8780.44
2026.01
79.5480.13
2025.05
79.3780.38
2026.01
79.2478.08
2025.05
79.275.53
2026.06
78.379
2025.05
78.1779.13
2026.01
78.1378.54
2025.05
78.0779.03
2025.05
78.0777.8
2025.05
78.0379.13
2025.05
77.8377.12
2026.03
77.3179.57
2026.03
77.2880.12
2026.01
77.2677.58
2026.03
77.1580.63
2025.05
76.2372.98
2025.05
76.1780.29
2026.06
75.878.5
2026.06
75.677.6
2026.03
75.4779.83
2025.05
75.379.75
2025.05
75.279.68
2025.05
74.579.29
2025.05
74.4374.53
2025.05
73.7378.6
2025.05
70.0776.43
2025.05
69.1375.91
2025.05
68.975.77
2025.05
68.3375.5
2025.05
68.175.15