通过对齐训练与推理路由器来稳定MoE强化学习

摘要

强化学习已成为提升大语言模型能力的关键方法。然而在混合专家模型中，路由机制常引发训练不稳定性，甚至导致灾难性的强化学习训练崩溃。我们分析了MoE模型的训练-推理一致性，发现两个阶段的路由行为存在显著差异。即使在相同条件下，路由框架在多次前向传播中也可能产生不同的专家选择。针对这一根本性不一致问题，我们提出滚动路由重放法：通过记录推理引擎的路由分布并在训练阶段重放，该方法显著降低了训练-推理策略的KL散度，在保持训练速度的同时有效缓解了极端差异。多场景实验表明，R3成功稳定了强化学习训练，避免了崩溃现象，其性能优于GSPO和TIS等方法。我们相信这项工作能为MoE模型的强化学习稳定性提供新的解决方案。

English

Reinforcement learning (RL) has emerged as a crucial approach for enhancing the capabilities of large language models. However, in Mixture-of-Experts (MoE) models, the routing mechanism often introduces instability, even leading to catastrophic RL training collapse. We analyze the training-inference consistency of MoE models and identify a notable discrepancy in routing behaviors between the two phases. Moreover, even under identical conditions, the routing framework can yield divergent expert selections across repeated forward passes. To address this foundational inconsistency, we propose Rollout Routing Replay (R3), a method that records routing distributions from the inference engine and replays them during training. R3 significantly reduces training-inference policy KL divergence and mitigates extreme discrepancies without compromising training speed. Extensive experiments on various settings confirm that R3 succeeds in stabilizing RL training, preventing collapse and outperforming methods such as GSPO and TIS. We believe this work can offer a new solution for stabilizing RL in MoE models.

通过对齐训练与推理路由器来稳定MoE强化学习

Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers

摘要

Support