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

Object Hallucination Evaluation on A-OKVQA POPE (Popular)

90.3Accuracy

SECOND

58.923267.069175.21583.3609Jun 10, 2025Jul 15, 2025Aug 19, 2025Sep 23, 2025Oct 28, 2025Dec 2, 2025Jan 7, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2025.06
90.3-90.590.4
2025.06
89.9-87.689.6
2025.06
89.9-92.590.7
2025.06
89.3-85.688.9
2025.06
89.3-85.489.1
2025.06
89.1-86.788.8
2025.06
89-86.788.7
2025.06
88.7-88.288.7
2025.06
88.7-83.888.1
2025.06
88.4-86.888.2
2025.06
88.2-8888.3
2025.06
88-88.388.1
2025.12
87.7190.96-87.26
2025.12
86.8--87.2
2025.12
86.589.59-85.95
2025.12
86.4790.74-86.52
2025.12
86.3688.73-86.2
2025.06
86.3-87.786.5
2025.12
86.2387.3-86.03
2025.12
85.9--84.7
2025.06
85.6-80.184.8
2025.12
85--85.1
2025.12
84.8--86.7
2025.12
84.7--84.3
2025.12
84.687.99-83.88
2025.12
84.5--83.3
2025.12
84.3--85
2025.06
84-7782.8
2025.06
82.1-79.181.6
2025.12
81.6--82.9
2025.12
81.4--80.8
2025.12
81--82.2
2025.12
80.8--81.9
2025.12
80.7976.29-83.76
2025.12
80.4775.61-82.35
2025.12
80.1172.64-82.36
2025.12
79.9--82.5
2025.12
79.5--82.6
2025.12
79.5--81.8
2025.12
79.0772.11-81.09
2025.12
78.8371.99-81.68
2025.12
78.873.38-81
2025.12
78.7372.83-81.17
2025.12
78.7--81.7
2025.12
78.6373.53-80.72
2025.12
77.870.98-80.91
2025.12
76.6369.59-80.19
2025.12
75.1770.15-77.91
2025.12
75.0768.58-78.77
2026.01
61.161.956.9359.22
2026.01
60.462.0653.5357.48
2026.01
60.1360.8656.858.76