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

Multimodal-Mind2Web

Benchmarks

Task NameDataset NameSOTA ResultTrend
GUI NavigationMultimodal-Mind2Web Cross-Website
Step Success Rate62.2
37
Web NavigationMultimodal-Mind2Web Cross-Domain
Element Accuracy57.4
15
Web NavigationMultimodal-Mind2Web Cross-Website
Element Accuracy60.5
15
Web NavigationMultimodal-Mind2Web Cross-Task
Element Accuracy60.8
15
Showing 4 of 4 rows