Which cross-embodiment mix would you pretrain on?
Open X-Embodiment, DROID and BridgeData are the usual building blocks, with LIBERO as the common simulated benchmark.
How do you weight them, and did adding your own robot's data early help more than fine-tuning later?
Referenced repos
Indexed from Hugging Face and linked upstream, not hosted here.
Unofficial community mirror of the Open X-Embodiment collection: about 55 robot sub-datasets repackaged as tar shards, together covering more than a million real trajectories.
LeRobot v3 conversion of DROID: about 95k real Franka teleoperation episodes recorded in many different scenes, with around 50k distinct language instructions.
LeRobot v2.0 conversion of BridgeData V2: about 53k WidowX tabletop and toy-kitchen episodes paired with roughly 20k distinct language instructions.
LeRobot-format copy of the LIBERO simulated Franka Panda benchmark, with about 1.7k demonstrations across its 40 language-conditioned tasks.