D3P learns to adaptively allocate denoising steps per robot action, matching fixed-step diffusion policy performance at 2.2x lower inference cost.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning
D3P learns to adaptively allocate denoising steps per robot action, matching fixed-step diffusion policy performance at 2.2x lower inference cost.