A little misleading. The title makes it seem like the robot learnt to write these unknown langauges but it looks like it was shown an image and was able to reproduce what was on the image, even though some of the writing was not in its training domain. Pre-LLMs so still pretty neat, just less impressive than the title suggests.
It doesn't seem to me like they trained it to write Japanese characters, but more broadly to reproduce a drawing or pencil strokes, so I don't think it is extremely surprising.
"Their learning system is split into two distinct models. A “local” model is in charge of what’s going on with the current stroke of the pen—so aiming in the right direction and determining how to end the stroke. And a “global” model is in charge of moving the robot’s writing utensil to the next stroke of the character."
From the paper: "To generate sufficient amount of training data for our model in simulation, we use KanjiVG [5], a database of Japanese Kanji characters. This database consists of both images of the character as well as an ordered list of strokes in the SVG format."
"Their learning system is split into two distinct models. A “local” model is in charge of what’s going on with the current stroke of the pen—so aiming in the right direction and determining how to end the stroke. And a “global” model is in charge of moving the robot’s writing utensil to the next stroke of the character."
That's the point.