نسخة أولية وصول مفتوح
Commit While Futures Agree: Consequence-Aware Adaptive Action Chunking for Robot Manipulation
Action-chunking policies predict multi-step control sequences, but a fundamental question remains: how much of a predicted action chunk should be committed before replanning? Existing systems typically execute a fixed-length prefix, implicitly assuming that the same execution horizon remains trustworthy across states. …