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

Object Hallucination on POPE

90.51Accuracy

Qwen3-VL-235B

77.593280.946684.387.6534May 9, 2026May 10, 2026May 12, 2026May 14, 2026May 16, 2026May 18, 2026May 20, 2026
Updated 13d ago

Evaluation Results

MethodLinks
2026.05
90.51-
2026.05
90.1-
2026.05
89.96-
2026.05
89.82-
2026.05
89.889.62
2026.05
89.77-
2026.05
89.68-
2026.05
89.4589.15
2026.05
89.2788.91
2026.05
89.1988.82
2026.05
89.04-
2026.05
89.03-
2026.05
88.94-
2026.05
88.92-
2026.05
88.9289.08
2026.05
88.74-
2026.05
88.6988.51
2026.05
88.54-
2026.05
88.3187.95
2026.05
88.1687.81
2026.05
87.68-
2026.05
87.54-
2026.05
87.53-
2026.05
87.31-
2026.05
87.12-
2026.05
87.1-
2026.05
87.02-
2026.05
87.01-
2026.05
87-
2026.05
86.92-
2026.05
86.84-
2026.05
86.83-
2026.05
86.77-
2026.05
86.54-
2026.05
86.5-
2026.05
86.28-
2026.05
85.7686.12
2026.05
85.6386.38
2026.05
85.4185.4
2026.05
85.27-
2026.05
85.1385.54
2026.05
84.8285.87
2026.05
84.74-
2026.05
84.7485.31
2026.05
84.7485.45
2026.05
84.685.28
2026.05
84.585.22
2026.05
84.1985.06
2026.05
83.0384.43
2026.05
80.22-
2026.05
78.09-