重访在线蒸馏策略：实证失效模式与简易修正方案

摘要

同策略蒸馏（OPD）因其能基于学生模型生成的轨迹而非固定的教师轨迹进行反馈评估，在大语言模型（LLM）后训练中备受关注。然而在长序列场景下，常用的逐词采样变体存在脆弱性：它将分布匹配简化为单步信号，且随着生成轨迹偏离教师模型的常见前缀，其可靠性会逐渐降低。我们从估计器设计与实现层面重新审视了OPD方法。理论分析表明，逐词级别的OPD相对于序列级反向KL散度存在偏差，但其最坏情况方差边界更紧；我们的模拟实验验证了该权衡关系——未来奖励关联性越强，梯度方差越大，学习稳定性越差。实证研究中，我们识别出采样词OPD的三大失效模式：失衡的单步信号、学生生成前缀上不可靠的教师指导，以及分词器或特殊词符失配导致的失真。针对这些问题，我们提出教师Top-K局部支持匹配方法，通过带截断的反向KL散度实现，结合Top-P轨迹采样与特殊词符掩码技术。在单任务数学推理与多任务（智能体+数学）训练中，该目标函数相比采样词OPD实现了更稳定的优化效果与更优的下游性能。

English

On-policy distillation (OPD) is appealing for large language model (LLM) post-training because it evaluates teacher feedback on student-generated rollouts rather than fixed teacher traces. In long-horizon settings, however, the common sampled-token variant is fragile: it reduces distribution matching to a one-token signal and becomes increasingly unreliable as rollouts drift away from prefixes the teacher commonly visits. We revisit OPD from the estimator and implementation sides. Theoretically, token-level OPD is biased relative to sequence-level reverse-KL, but it has a much tighter worst-case variance bound; our toy study shows the same tradeoff empirically, with stronger future-reward coupling producing higher gradient variance and less stable learning. Empirically, we identify three failure modes of sampled-token OPD: an imbalanced one-token signal, unreliable teacher guidance on student-generated prefixes, and distortions caused by tokenizer or special-token mismatch. We address these issues with teacher top-K local support matching, implemented as truncated reverse-KL with top-p rollout sampling and special-token masking. Across single-task math reasoning and multi-task agentic-plus-math training, this objective yields more stable optimization and better downstream performance than sampled-token OPD.

重访在线蒸馏策略：实证失效模式与简易修正方案

Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes

摘要

Support