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

Step success rate on AndroidControl (IDD)

69.4Step Success Rate (SSR)

T3A* (Qwen2.5-7B-Instruct-ASL)

Updated 11d ago

Evaluation Results

Method	Links
T3A* (Qwen2.5-7B-Instruct-ASL) 2025.06		69.4
T3A* (Qwen2.5-7B-Instruct-SFT) 2025.06		67.4
M3A (GPT-4o) 2025.06		60.8
T3A (GPT-4o) 2025.06		56.1
M3A (Gemini-2.5-Flash) 2025.06		49.4
T3A (Gemini-2.5-Flash) 2025.06		49.1
T3A* (Gemini-2.5-Flash) 2025.06		46.8
SeeAct (Gemini-2.5-Flash) 2025.06		36.7
SeeAct (GPT-4o) 2025.06		31.5
T3A* (Qwen2.5-7B-Instruct) 2025.06		26.1