List your company

Web Task Completion on MiniWoB++ Held-out Tasks (test)

70.4Success RateClaude-3.5-Sonnet
35.0444.2253.462.58Jul 5, 2025

Evaluation results

MethodLinks
Claude-3.5-SonnetBackbone Model=Claude-3.5-Sonnet, Training Protocol=Instruct2025.0770.4
GPT-4o-MiniBackbone Model=GPT-4o-Mini, Training Protocol=Instruct2025.0766.1
o1-MiniBackbone Model=o1-Mini, Training Protocol=Instruct2025.0766.1
Llama-3.1-70B-InstructBackbone Model=Llama-3.1-70B, Training Protocol=Instruct2025.0765.2
Llama-3.1-405B-InstructBackbone Model=Llama-3.1-405B, Training Protocol=Instruct2025.0765.2
GPT-4oBackbone Model=GPT-4o, Training Protocol=Instruct2025.0764.3
Llama-3.1-8B SFT+RL (Ours)Backbone Model=Llama-3.1-8B, Training Protocol=SFT+RL2025.0763.3
Llama-3.3-70B Instruct (Teacher)Backbone Model=Llama-3.3-70B, Training Protocol=Instruct, Model Role=Teacher2025.0761.9
Qwen-2.5-7B SFT+RL (Ours)Backbone Model=Qwen-2.5-7B, Training Protocol=SFT+RL2025.0761.5
Qwen-2.5-72B Instruct (Teacher)Backbone Model=Qwen-2.5-72B, Training Protocol=Instruct, Model Role=Teacher2025.0759
Llama-3.1-8B SFT (Ours)Backbone Model=Llama-3.1-8B, Training Protocol=SFT2025.0756.7
Qwen-2.5-7B SFT (Ours)Backbone Model=Qwen-2.5-7B, Training Protocol=SFT2025.0756.5
Qwen-2.5-7B RL (Ours)Backbone Model=Qwen-2.5-7B, Training Protocol=RL2025.0753
Llama-3.1-8B RL (Ours)Backbone Model=Llama-3.1-8B, Training Protocol=RL2025.0743.5
Qwen-2.5-7B Instruct (Student)Backbone Model=Qwen-2.5-7B, Training Protocol=Instruct, Model Role=Student2025.0737
Llama-3.1-8B Instruct (Student)Backbone Model=Llama-3.1-8B, Training Protocol=Instruct, Model Role=Student2025.0736.4