When Should the Teacher Move? Temporal Coupling and Stability in Self On-Policy Distillation
arXiv:2606.03532v2 Announce Type: replace-cross Abstract: Self on-policy distillation trains a student policy against a teacher derived from its own parameter history, yet the teacher's update schedule -- which governs the \emph{temporal coupling} between teacher and student -- has not been systematically studied as a stability variable. Through a controlled schedule sweep on Qwen3-8B, we establish that \emph{isolation periods}, defined as complete teacher freezing between updates, are the key…