The minecraft assistant is pretty exciting. I haven’t read the paper and would accept “read the paper” as an answer, but curious about details like “what are examples of tasks the helper-bot does?” and “how does the human interact with the bot, in-game or out-of-game.”
Also I vaguely remember that you had some kind of minecraft AI project and curious if this was related to that at all.
So far it is more of a position paper saying “we should do this” rather than “we have done this”, so it’s not super clear what tasks the helper-bot does. The human interacts with the bot by messaging it in normal English. Here’s an example of the kind of thing they want to be able to do, taken directly from the paper:
PLAYER: build a tower 15 blocks tall and then put a giant smiley on top
ASSISTANT: ok [assistant starts building the tower]
PLAYER: wait, stop, make every other block red [assistant recognizes the instruction refers to a change in its current build task, but doesn’t understand the change (and specifically recognizes “every other” as unknown)]
ASSISTANT: What is “every other”?
PLAYER: Let me show you [player makes a stack of blocks alternating in color; assistant is able to generalize “every other” to new situations]
My project is not at all related in the sense that this paper was a surprise to me when it was released a few days ago, but it is very related in the sense that I too am trying to build a benchmark of Minecraft tasks where there aren’t obvious reward functions. I’m less focused on natural language though—the hope is that people could try to solve it using other techniques as well, such as IRL or imitation learning.
The minecraft assistant is pretty exciting. I haven’t read the paper and would accept “read the paper” as an answer, but curious about details like “what are examples of tasks the helper-bot does?” and “how does the human interact with the bot, in-game or out-of-game.”
Also I vaguely remember that you had some kind of minecraft AI project and curious if this was related to that at all.
So far it is more of a position paper saying “we should do this” rather than “we have done this”, so it’s not super clear what tasks the helper-bot does. The human interacts with the bot by messaging it in normal English. Here’s an example of the kind of thing they want to be able to do, taken directly from the paper:
My project is not at all related in the sense that this paper was a surprise to me when it was released a few days ago, but it is very related in the sense that I too am trying to build a benchmark of Minecraft tasks where there aren’t obvious reward functions. I’m less focused on natural language though—the hope is that people could try to solve it using other techniques as well, such as IRL or imitation learning.