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

Object Hallucination Evaluation on POPE A-OKVQA

89.23Accuracy

VLI

64.571670.973377.37583.7767Oct 15, 2025Nov 6, 2025Nov 29, 2025Dec 22, 2025Jan 13, 2026Feb 5, 2026Feb 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
89.2388.79
2025.10
89.0789.63
2026.02
89.0388.28
2025.10
89.0289.52
2025.10
8989.59
2025.10
88.6889.25
2026.02
88.1387.32
2026.02
88.187.27
2026.02
87.788.8
2026.01
87.6887.33
2026.02
87.6387.06
2026.01
87.5387.18
2026.02
87.3788.55
2026.02
87.3388.52
2026.01
87.1386.82
2026.02
87.0788.29
2026.02
87.0786.37
2026.01
86.9386.38
2026.01
86.5385.83
2026.02
86.5387.93
2026.02
86.2785.56
2026.02
86.285.45
2026.01
85.8785.54
2026.02
85.8385.32
2026.02
8586.52
2026.01
84.9383.53
2026.02
84.6786.53
2026.01
84.5884.23
2026.02
84.4385.91
2025.10
84.2584.13
2026.01
84.1383.77
2026.01
83.8783.54
2026.01
83.0282.54
2026.02
8384.66
2025.10
82.883.2
2025.10
82.3483.08
2026.01
82.1482.93
2025.10
81.9382.16
2026.01
81.4479.82
2025.10
80.9780.79
2026.02
80.9381.6
2025.10
80.8681.87
2026.02
80.7787.86
2026.02
80.5783.39
2026.02
80.5383.32
2025.10
80.5181.11
2026.02
80.4783.33
2026.01
80.4380.92
2026.02
80.383.22
2026.02
80.381.09
2026.02
80.180.93
2026.02
79.9783.04
2025.10
79.8179.86
2026.01
79.5479.81
2026.02
79.5380.73
2025.10
78.9878.02
2025.10
78.8577.73
2025.10
78.2376.78
2026.02
77.4781.39
2026.02
77.4381.02
2025.10
76.6474.68
2026.02
76.5780.91
2026.02
74.3785.48
2025.10
72.0975.37
2026.02
72.0380.73
2025.10
70.0173.4
2026.02
69.7376.32
2026.02
69.5376.22
2026.02
69.4776.18
2026.02
69.3776.12
2025.10
69.2672.23
2026.02
68.9775.96
2026.02
67.9775.05
2026.02
67.0374.93
2025.10
65.5268.36