Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Explanation on T-IC13 adapted (test)
Loading...
86.7
CSS
DocShield
35.22
48.585
61.95
75.315
Apr 3, 2026
CSS
BS(F1)
Updated 13d ago
Evaluation Results
Method
Method
Links
CSS
BS(F1)
DocShield
Zero-shot=true
2026.04
86.7
78.5
Gemini-2.5-Pro
Zero-shot=true
2026.04
76.9
68.7
qwen3-vl-8B
Zero-shot=true
2026.04
72.1
66.5
InternVL-3.5-8B
Zero-shot=true
2026.04
71.4
66.4
gpt-4o-o3-0416
Zero-shot=true
2026.04
69.1
65.7
Qwen2.5-VL-7B
Zero-shot=true
2026.04
64.4
65.5
FakeShield
Zero-shot=true
2026.04
57.5
52.7
DeepSeekVL-7B
Zero-shot=true
2026.04
37.2
28.9
Feedback
Search any
task
Search any
task