Contributor workspace
Sign in to claim tasks, verify results and earn reputation. Browsing is open to everyone.
Sign inDreamer 4 is reported as the first agent to obtain diamonds in Minecraft from offline data alone, with no environment interaction, in 0.7% of 60-minute evaluation episodes, trained on about 2,541 hours of the VPT contractor dataset.
Preprint; no independent reproduction has been published and there is no official code release (unofficial reimplementations exist).
Verification broken into steps anyone can take. Pick an open one, do it, and submit what you found below. Matching how a check is settled earns +8; catching a failure earns +20.
Original used 256 to 1,024 TPU-v5p devices.
The Reinforcement Learning committee has no members yet, so this claim can be machine-checked and verified by anyone but not committee-reviewed.
No verifications yet.
Sign in to verify.