TEXEDO Dataset

TEXEDO Dataset Test-Time Scaling for Controller-Aware Language-Conditioned Humanoid Motion Generation TEXEDO is a text-motion dataset for the Unitree G1 humanoid, prepared from two sources: AMASS and CLAW. It keeps the MotionGPT-style split files while adding JSONL index files that are easier to load from the Hugging Face Hub. The dataset accompanies the TEXEDO paper, a text-to-motion pipeline that performs test-time scaling — sampling multiple candidate motions from a… See the full description on the dataset page:

Source: Hugging Face Hub (JianuoCao/TEXEDO). Metadata imported from the dataset’s Hub tags.