DARP reparameterizes imitation learning around local neighborhood structure using k-NN expert states, actions, and relative distance vectors, delivering 15-46% gains over behavior cloning in control and manipulation tasks.
arXiv preprint arXiv:2507.21452 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Difference-Aware Retrieval Policies for Imitation Learning
DARP reparameterizes imitation learning around local neighborhood structure using k-NN expert states, actions, and relative distance vectors, delivering 15-46% gains over behavior cloning in control and manipulation tasks.