You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: src/agentlab/agents/tool_use_agent/hint_db.csv
+5-1Lines changed: 5 additions & 1 deletion
Original file line number
Diff line number
Diff line change
@@ -12,4 +12,8 @@ June 11,miniwob.drag-items-grid,30,claude-3-7-sonnet-20250219,MultiToolUse-claud
12
12
June 11,miniwob.drag-items-grid,30,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,aman,Dragging items,"Hovering near a target item can prompt it to move to an adjacent grid cell. Use this behavior to clear the destination before dropping the dragged item"
13
13
June 11,miniwob.drag-items-grid,30,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,aman,Dragging items,"Move items along slightly curved instead of straight lines to better imitate human-like dragging behavior."
14
14
June 11,miniwob.drag-items,30,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,aman,Dragging items,"Move items along slightly curved instead of straight lines to better imitate human-like dragging behavior."
15
-
June 11,miniwob.drag-items,30,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,aman,Dragging items,"For dragging tasks, use only mouse_down, move_mouse, and mouse_up. Do not perform a single continuous drag. Instead, move the mouse in steps using move_mouse, and then release with mouse_up when the target is reached."
15
+
June 11,miniwob.drag-items,30,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,aman,Dragging items,"For dragging tasks, use only mouse_down, move_mouse, and mouse_up. Do not perform a single continuous drag. Instead, move the mouse in steps using move_mouse, and then release with mouse_up when the target is reached."
16
+
June 18,miniwob.count-shape,23,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,allac,Shape and letters size comparison in miniwob,"Shapes or items have different colors and different size. Size is relative to the other objects in the white area and is either ""large"" or ""small"". Shapes that are larger than the average shape or letter are considered ""large"". Others are ""small""."
17
+
June 18,miniwob.count-shape,23,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,allac,communicate answer in miniwob,Answer by clicking one of the buttons describing multiple choices.
18
+
June 18,miniwob.count-shape,23,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,allac,Simbols of colors in miniwob,"Colors a distinct in this task, e.g., cyan is not a type of blue. "
19
+
June 18,miniwob.form-sequence-2,23,claude-3-7-sonnet-20250219,MultiToolUse-claude-3-7-sonnet-20250219,miniwob,miniwob,allac,Reporting results in miniwob,Make sure to click submit to finish the task.
0 commit comments