Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Our Benchmark

Benchmarks

Task NameDataset NameSOTA ResultTrend
Differential DiagnosisOur Benchmark
Diagnostic Accuracy45.26
14
Image DescriptionOur Benchmark
METEOR23.64
14
AI-generated image detectionOur benchmark In-the-Wild
Accuracy96.7
5
AI-generated image detectionOur benchmark In-the-Lab
Accuracy97.4
5
AI-generated image detectionOur benchmark (Overall)
Accuracy97.05
5
Indirect injection interceptionOur Benchmark indirect EN n=180
IR95.6
2
Showing 6 of 6 rows