List your company

Web Task on Mind2Web (test)

31.3Step-wise Success RateLUMOS-I_Web-13B
1.148.9716.824.63Nov 9, 2023

Evaluation results

MethodLinks
LUMOS-I_Web-13BType=Open-source, Fine-tuned=true, Backbone Size=13B2023.1131.3
LUMOS-I_WebType=Open-source, Fine-tuned=true2023.1127.6
GPT-4Type=GPT/API-based2023.1122.6
GPT-3.5-turboType=GPT/API-based2023.1115.7
AgentLM-70BType=Open-source, Fine-tuned=true2023.1113.5
Koala-13BType=Open-source2023.116
WizardLM-30BType=Open-source2023.113.1
Baichuan-13B-chatType=Open-source2023.112.3