Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Grounding on RealText proposed (test)
Loading...
32.4
mIOU
DocShield
4.008
11.379
18.75
26.121
Apr 3, 2026
mIOU
mF1
Updated 13d ago
Evaluation Results
Method
Method
Links
mIOU
mF1
DocShield
2026.04
32.4
36.7
FakeShield
2026.04
30.5
12.4
Qwen2.5-VL-7B
2026.04
17.8
15.5
gpt-4o-o3-0416
2026.04
8.1
15.1
InternVL-3.5-8B
2026.04
7.4
9.4
qwen3-vl-8B
2026.04
7.3
10.7
Gemini-2.5-Pro
2026.04
5.3
10.8
DeepSeekVL-7B
2026.04
5.1
6.7
Feedback
Search any
task
Search any
task