Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Object Hallucination Detection on POPE averaged across MS-COCO, A-OKVQA, and GQA (Adversarial)
Loading...
0.807
Accuracy
VAF
0.7706
0.78005
0.7895
0.79895
Mar 17, 2025
Accuracy
F1 Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
F1 Score
VAF
Backbone=LLaVA-v1.5-13B
2025.03
0.807
0.817
VAF
Backbone=Qwen-VL-Chat-7B
2025.03
0.804
0.812
VAF
Backbone=LLaVA-v1.5-7B
2025.03
0.801
0.81
ICD
Backbone=LLaVA-v1.5-13B
2025.03
0.791
0.801
VCD
Backbone=Qwen-VL-Chat-7B
2025.03
0.788
0.801
ICD
Backbone=LLaVA-v1.5-7B
2025.03
0.785
0.799
VCD
Backbone=LLaVA-v1.5-13B
2025.03
0.782
0.797
VCD
Backbone=LLaVA-v1.5-7B
2025.03
0.781
0.796
ICD
Backbone=Qwen-VL-Chat-7B
2025.03
0.781
0.792
Regular
Backbone=LLaVA-v1.5-13B
2025.03
0.778
0.795
Regular
Backbone=LLaVA-v1.5-7B
2025.03
0.776
0.794
Regular
Backbone=Qwen-VL-Chat-7B
2025.03
0.772
0.789
Feedback
Search any
task
Search any
task