Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Adversarial Figure Manipulation Detection on PaperGuard Visual Attacks
Loading...
93.5
Accuracy
Chunk-based
33.284
48.917
64.55
80.183
Jun 10, 2026
Accuracy
Recall
False Positive Rate (FPR)
False Negative Rate (FNR)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Recall
False Positive Rate (FPR)
False Negative Rate (FNR)
Chunk-based
Retrieval Type=image-c...
2026.06
93.5
90.32
10
9.68
Embedding Classifier
2026.06
72.5
0
10
1
LLM-as-a-Judge
2026.06
35.6
90
100
10
Feedback
Search any
task
Search any
task