חסרות לוגיטים ישנים בRL אסינכרוני: תאושטות סמנטית ושיטות תיקון לתיקון-לא-פוליצי

תקציר מקורי באנגליתarXiv:2605.12070v3 Announce Type: replace-cross Abstract: Asynchronous reinforcement learning improves rollout throughput for large language model agents by decoupling sample generation from policy optimization, but it also introduces a critical failure mode for PPO-style off-policy correction. In heterogeneous training systems, the total importance ratio should ideally be decomposed into two semantically distinct factors: a \emph{training--inference discrepancy term} that aligns inference-side and training-side distributions at the same behavior-policy version, and a \emph{policy-staleness term} that constrains the update from the historical policy to the current policy. We show that practical asynchronous pipelines with delayed updates and partial rollouts often lose the required histori