Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Object Hallucination Probing on A-OKVQA (Random split)
Loading...
90.83
Accuracy
SchroMind
83.1548
85.1474
87.14
89.1326
Feb 10, 2026
Accuracy
F1 Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
F1 Score
SchroMind
Model=LLaVA-1.5-7B
2026.02
90.83
90.77
Scalpel
Base Model=LLaVA-1.5-7B
2026.02
89.87
89.93
ICT
Base Model=LLaVA-1.5-7B
2026.02
89.3
89.4
ICT
Model=LLaVA-1.5-7B
2026.02
89.3
89.4
OPERA
Base Model=LLaVA-1.5-7B
2026.02
88.02
84.59
OPERA
Model=LLaVA-1.5-7B
2026.02
88.02
84.59
VCD
Base Model=LLaVA-1.5-7B
2026.02
86.15
86.34
VCD
Model=LLaVA-1.5-7B
2026.02
86.15
86.34
AVISC
Model=LLaVA-1.5-7B
2026.02
84.6
85.88
M3ID
Model=LLaVA-1.5-7B
2026.02
83.57
85.09
Vanilla
Base Model=LLaVA-1.5-7B
2026.02
83.45
82.56
Regular
Model=LLaVA-1.5-7B
2026.02
83.45
82.56
Feedback
Search any
task
Search any
task
