MoE-DAgger: Transition-Aware Router-Entropy Gating for Selective Human Correction in Phase-Structured Manipulation
MoE-DAgger: Transition-Aware Router-Entropy Gating for Selective Human Correction in Phase-Structured Manipulation
This paper introduces MoE-DAgger, a method that uses causal router history to distinguish normal phase transitions from errors in robot manipulation without explicit phase labels. The approach achieves 84.0% success on the Green task and 78.0% on the Red task under matched human-control budgets, outperforming phase-unaware filtered entropy by 22 percentage points on Green and 26 points on Red. No collisions were observed in 50 rollouts per task.
The study presents MoE-DAgger, a method using causal router history to distinguish normal phase transitions from errors in robot manipulation without explicit phase labels, achieving 84.0% success on the Green task and 78.0% on the Red task under matched human-control budgets.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
atched active human-control budgets, MoE-DAgger achieves 84.0% task success on Green and 78.0% on Red, exceeding phase-unaware filtered entropy by 22 percentage points on Green and 26 points on Red, with no observed collisions in 50 rollouts per task. On 50 shared intervention-free traces per task, 83.3% and 80.0% of its first proposals fall on failed rollouts within predefined failure windows when 15 and 25 traces a
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org