osunlp/Mind2Web
Viewer
•
Updated
•
253
•
733
•
100
Towards Generalist Agents for the Web
Note Original Mind2Web Dataset
Note Multimodal version of the Mind2Web Dataset
Note An Online benchmark of Mind2Web-level web tasks
Note First LLM-based web agent Ecologically valid eval
Note SeeAct: First generalist web agent with visual perception
View and refresh evaluation leaderboards for both human and automated tasks
Note Leaderboard of Online-Mind2Web