Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Referring Text Classification on COCO-Text Out-of-Domain
Loading...
84.62
Accuracy
LLaVA-1.5 + Blur
20.66
37.265
53.87
70.475
Feb 23, 2026
Accuracy
Updated 2d ago
Evaluation Results
Method
Method
Links
Accuracy
LLaVA-1.5 + Blur
Input format=Box, Trai...
2026.02
84.62
LLaVA-1.5 + Blur
Input format=Mask, Tra...
2026.02
75.85
LLaVA-1.5 + ControlMLLM++
Input format=Box, Trai...
2026.02
74.66
LLaVA-1.5 + ControlMLLM++
Input format=Mask, Tra...
2026.02
74.65
LLaVA-1.5 + ControlMLLM
Input format=Box, Trai...
2026.02
63.06
LLaVA-1.5 + ControlMLLM
Input format=Mask, Tra...
2026.02
61.43
Shikra-7B
Input format=Box, Trai...
2026.02
60.41
Ferret-7B
Input format=Mask, Tra...
2026.02
59.81
LLaVA-1.5 + Color
Input format=Box, Trai...
2026.02
59.04
Ferret-7B
Input format=Box, Trai...
2026.02
58.7
LLaVA-1.5
Input format=Box, Trai...
2026.02
57.42
LLaVA-1.5
Input format=Mask, Tra...
2026.02
57.42
GPT4-ROI
Input format=Box, Trai...
2026.02
57.17
LLaVA-1.5 + Color
Input format=Mask, Tra...
2026.02
56.57
LLaVA-1.5 + Edit Att
Input format=Mask, Tra...
2026.02
51.88
LLaVA-1.5 + Edit Att
Input format=Box, Trai...
2026.02
50
Kosmos-2
Input format=Box, Trai...
2026.02
23.12
Feedback
Search any
task
Search any
task