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

AITZ

Benchmarks

Task NameDataset NameSOTA ResultTrend
GUI Task ExecutionAITZ
Success Rate66.64
20
GUI AgentAITZ
SR69.4
20
GUI Agent Navigation and ActionAITZ
Success Rate (SR)69.15
18
Action PredictionAITZ (test)
Scroll Accuracy76.12
17
Action PredictionAITZ
Click TMR95.9
15
Action Prediction AttackAITZ
Privacy Attack TMR100
12
Mobile Agent UI InteractionAITZ Benchmark
Scroll Rate88.12
11
Mobile GUI Task ExecutionAITZ zero-shot
Scroll Rate91.17
8
Mobile GUI AutomationAITZ
Type Success Rate64.44
7
GUI Action PredictionAITZ
Task Match (TM)77.05
5
GUI NavigationAITZ 1.0 (test)
Scroll SR0.9117
5
UI OperationsAITZ
Task Metric (TM)89.6
3
Showing 12 of 12 rows