List your company

GUI Agent Robustness Evaluation on Pop-up Perturbation 1.0 (Medium)

97Direct Success Rate (DSR)CA
6.5230.0153.576.99Jul 13, 2025

Evaluation results

MethodLinks
CABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Default2025.07971003
CABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.0796.51003.5
CABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Default2025.07931007
CABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0792.51007.5
CABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.079285.46.6
CABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0786.484.91.5
CABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.07789416
CABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Default2025.07729321
NDBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0771.484.413
NDBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0770.980.49.5
NDBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.0767.58315.5
NDBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Default2025.0759.578.519
DPOBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0757.378.921.6
DPOBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0756.875.418.6
DPOBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.074775.528.5
DPOBackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Default2025.0745.57327.5
DABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.074579.534.5
DABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0743.265.822.6
DABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Default2025.0742.583.541
DABackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0741.766.324.6
DABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.0741.58038.5
DABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Inductive, Visual Prompting=Default2025.074179.538.5
DABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0732.778.445.7
DABackbone=LLaVA-v1.6-Vicuna-13B (LaSM L12-28, α=1.2), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0732.778.445.7
NDBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0720.668.848.2
NDBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0720.172.952.8
DPOBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Default2025.0717.665.848.2
DPOBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Overlay, Visual Prompting=Highlight2025.0717.665.347.7
NDBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Default2025.071569.554.5
NDBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.0713.569.556
DPOBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Default2025.071065.555.5
DPOBackbone=Qwen2-vl-7B (LaSM L7-18, α=1.1), Injection Type (IT)=Inductive, Visual Prompting=Highlight2025.07106656