RLPF uses reinforcement learning with a physics-simulator tracking reward and an alignment verification module to fine-tune a large text-to-motion model for physically feasible humanoid motions.
Since SMPL parameters represent various human body shapes, we first optimize the shape parameter β′ to approximate a humanoid form
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.RO 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
RL from Physical Feedback: Aligning Large Motion Models with Humanoid Control
RLPF uses reinforcement learning with a physics-simulator tracking reward and an alignment verification module to fine-tune a large text-to-motion model for physically feasible humanoid motions.