ASPIRE uses code-as-policy, closed-loop execution traces for autonomous repair, a distilling skill library, and evolutionary search to discover persistent robot skills that transfer across tasks, simulation, real robots, and embodiments, reporting large gains over priors on manipulation benchmarks.
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
-
ASPIRE: Agentic /Skills Discovery for Robotics
ASPIRE uses code-as-policy, closed-loop execution traces for autonomous repair, a distilling skill library, and evolutionary search to discover persistent robot skills that transfer across tasks, simulation, real robots, and embodiments, reporting large gains over priors on manipulation benchmarks.