WARP trains a reward model on time-warped successful demonstrations to produce frame-level progress estimates that upweight high-advantage chunks during behavior cloning, maintaining high success rates on suboptimal datasets where vanilla BC fails.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
WARP-RM: A Warp-Augmented Relative Progress Reward Model for Data Curation
WARP trains a reward model on time-warped successful demonstrations to produce frame-level progress estimates that upweight high-advantage chunks during behavior cloning, maintaining high success rates on suboptimal datasets where vanilla BC fails.